:: ghana statistical service :: - reflect the uk governments official … · 2016. 11. 21. · this...
TRANSCRIPT
This study was funded by UK aid from the UK government. However, the views expressed do not necessarily
reflect the UK government’s official policies.
Suggested citation:
Institute of Development Studies (IDS), Ghana Statistical Services (GSS) and Associates (2016), Domestic
Violence in Ghana: Incidence, Attitudes, Determinants and Consequences, Brighton: IDS.
© Institute of Development Studies 2016
ISBN: 978-1-78118-314-4
This is an Open Access report distributed under the terms of the Creative Commons Attribution 4.0
International licence, which permits unrestricted use, distribution, and reproduction in any medium,
provided the original authors and source are credited. http://creativecommons.org/licenses/by/4.0/legalcode
1
Contents
List of tables and figures 4
Foreword 7
Acknowledgements 9
Research team 12
Acronyms and abbreviations 14
Executive summary 15
1. Background to the study 20
1.1. Domestic violence in Ghana 21
1.2. The study 25
2. Conceptual framework 27
2.1. Types and perpetration of domestic violence 31
2.2. Attitudes towards domestic violence 34
2.3. Determinants of domestic violence 35
2.4. Consequences of domestic violence 39
2.5. Use of and access to victims’ support services and institutions 41
3. Methodology 44
3.1. Quantitative instruments 44
3.2. Qualitative instruments 46
3.3. Sampling design: quantitative survey 48
3.4. Sampling design: qualitative fieldwork 49
3.5. Training and pre-testing 50
3.6. Fieldwork implementation 51
3.7. Response rates 52
3.8. Data analysis 57
3.9. Ethical considerations 58
4. Empirical results 61
4.1. Incidence of violence (domestic and non-domestic) against women and men in Ghana 61
4.1.1. Social violence 62
4.1.2. Physical violence 73
4.1.3. Sexual violence 83
4.1.4. Psychological violence 93
4.1.5. Economic violence 103
4.2. Incidence of domestic violence against women and men in Ghana 113
4.2.1. Domestic social violence 114
4.2.2. Domestic physical violence 121
4.2.3. Domestic sexual violence 128
4.2.4. Domestic psychological violence 135
4.2.5. Domestic economic violence 142
4.2.6. Summary 149
2
4.3. Perpetration of domestic violence in Ghana 150
4.3.1. Violence perpetration within domestic relations 150
4.3.2. Profile of domestic violence perpetrators 153
4.3.3. Domestic violence against children 158
4.4. Attitudes towards domestic violence in Ghana 159
4.4.1. Attitudes towards domestic social violence 160
4.4.2. Attitudes towards domestic physical violence 166
4.4.3. Attitudes towards domestic sexual violence 179
4.4.4. Attitudes towards domestic psychological violence 182
4.4.5. Attitudes towards domestic economic violence 184
4.5. Determinants of domestic violence in Ghana 185
4.5.1. Age and intergenerational effects 187
4.5.2. Socio-economic effects 189
4.5.3. Marital status 193
4.5.4. Geographical effects 196
4.5.5. Patriarchal and gender norms 197
4.5.6. Exposure to other forms of violence 200
4.6. Consequences of domestic violence in Ghana 201
4.6.1. Physical health effects 202
4.6.2. Mental health and emotional effects 206
4.6.3. Effects on work and daily activities 208
4.6.4. Consequences for children 209
4.7. Use of and access to victims’ support services and institutions 211
4.7.1. Seeking external help 211
4.7.2. Public services and institutions 215
4.7.3. Knowledge about support services and institutions 219
4.7.4. Levels of satisfaction about support services and institutions 223
5. Summary and discussion 225
5.1. Incidence of domestic violence in Ghana 225
5.2. Attitudes towards domestic violence in Ghana 226
5.3. Determinants of domestic violence in Ghana 227
5.4. Consequences of domestic violence in Ghana 230
5.5. Use of and access to victims’ support services and institutions 231
6. Way forward and policy recommendations 233
6.1. Recommendations with respect to the incidence of domestic violence in Ghana 234
6.2. Recommendations with respect to attitudes towards domestic violence in Ghana 236
6.3. Recommendations with respect to the determinants of domestic violence in Ghana 238
6.4. Recommendations with respect to the consequences of domestic violence in Ghana 240
6.5. Recommendations with respect to the use of and access to victims’ support services and
institutions in Ghana 241
6.6. Recommendations for future research on domestic violence in Ghana 244
References 247
3
Appendices 259
Appendix A: Incidence of domestic violence worldwide 260
Appendix B: Timeline of major policies related to domestic and gender-based violence 263
Appendix C: Summary of qualitative sample, methods and instruments 265
Appendix D: Distribution of enumeration areas across regions and survey weights 266
Appendix E: Regression tables 269
Questionnaire annex 275
4
List of tables and figures
Table 1: Response rates to household survey (in percentages) .............................................. 53
Table 2: Response rates to the individual survey: women (in percentages) ........................ 54
Table 3: Response rates to the individual survey: men (in percentages) .............................. 54
Table 4: Distribution of survey respondents by key socio-economic characteristics
(weighted) ...................................................................................................................................... 55
Table 5: Incidence of violence in Ghana 2015 (in percentages) .............................................. 62
Table 6: Percentage of women who have experienced domestic or non-domestic social
violence over their lifetime .......................................................................................................... 65
Table 7: Percentage of women who experienced domestic or non-domestic social violence
in the last 12 months ..................................................................................................................... 67
Table 8: Percentage of men who have experienced domestic or non-domestic social
violence over their lifetime .......................................................................................................... 69
Table 9: Percentage of men who experienced domestic or non-domestic social violence in
the last 12 months .......................................................................................................................... 71
Table 10: Percentage of women who have experienced domestic or non-domestic physical
violence over their lifetime .......................................................................................................... 75
Table 11: Percentage of women who experienced domestic or non-domestic physical
violence in the last 12 months ..................................................................................................... 77
Table 12: Percentage of men who have experienced domestic or non-domestic physical
violence over their lifetime .......................................................................................................... 79
Table 13: Percentage of men who experienced domestic or non-domestic physical
violence in the last 12 months ..................................................................................................... 81
Table 14: Percentage of women who have experienced domestic or non-domestic sexual
violence over their lifetime .......................................................................................................... 85
Table 15: Percentage of women who experienced domestic or non-domestic sexual
violence in the last 12 months ..................................................................................................... 87
Table 16: Percentage of men who have experienced domestic or non-domestic sexual
violence over their lifetime .......................................................................................................... 89
Table 17: Percentage of men who experienced domestic or non-domestic sexual violence
in the last 12 months ..................................................................................................................... 91
Table 18: Percentage of women who have experienced domestic or non-domestic
psychological violence over their lifetime ................................................................................. 95
Table 19: Percentage of women who experienced domestic or non-domestic psychological
violence in the last 12 months ..................................................................................................... 97
Table 20: Percentage of men who have experienced domestic or non-domestic
psychological violence over their lifetime ................................................................................. 99
Table 21: Percentage of men who experienced domestic or non-domestic psychological
violence in the last 12 months ................................................................................................... 101
Table 22: Percentage of women who have experienced domestic or non-domestic
economic violence over their lifetime ...................................................................................... 105
Table 23: Percentage of women who experienced domestic or non-domestic economic
violence in the last 12 months ................................................................................................... 107
5
Table 24: Percentage of men who have experienced domestic or non-domestic economic
violence over their lifetime ........................................................................................................ 109
Table 25: Percentage of men who experienced domestic or non-domestic economic
violence in the last 12 months ................................................................................................... 111
Table 26: Incidence of domestic violence in Ghana 2015 in the 12 months prior to the
survey (in percentages) .............................................................................................................. 113
Table 27: Percentage of domestic social violence over the last 12 months ......................... 114
Table 28: Percentage of women who experienced domestic social violence over the last
12 months ..................................................................................................................................... 117
Table 29: Percentage of men who experienced domestic social violence over the last 12
months .......................................................................................................................................... 119
Table 30: Percentage of domestic physical violence over the last 12 months..................... 121
Table 31: Percentage of women who experienced domestic physical violence over the
last 12 months .............................................................................................................................. 124
Table 32: Percentage of men who experienced domestic physical violence over the last
12 months ..................................................................................................................................... 126
Table 33: Percentage of domestic sexual violence over the last 12 months ........................ 128
Table 34: Percentage of women who experienced domestic sexual violence over the last
12 months ..................................................................................................................................... 131
Table 35: Percentage of men who experienced domestic sexual violence over the last 12
months .......................................................................................................................................... 133
Table 36: Percentage of domestic psychological violence over the last 12 months ........... 135
Table 37: Percentage of women who experienced domestic psychological violence over
the last 12 months ........................................................................................................................ 138
Table 38: Percentage of men who experienced domestic psychological violence over the
last 12 months .............................................................................................................................. 140
Table 39: Percentage of domestic economic violence over the last 12 months .................. 142
Table 40: Percentage of women who experienced domestic economic violence over the
last 12 months .............................................................................................................................. 145
Table 41: Percentage of men who experienced domestic economic violence over the last
12 months ..................................................................................................................................... 147
Table 42: Socio-economic groups with highest incidence of domestic violence ................ 149
Table 43: Perpetrators of domestic violence as reported by women (in percentages) ...... 151
Table 44: Perpetrators of domestic violence as reported by men (in percentages) ........... 152
Table 45: Percentage and number of domestic violence perpetrators ................................. 153
Table 46: Number and percentage of perpetrators reporting violence against same or
another sex ................................................................................................................................... 154
Table 47: Percentage of women reporting perpetration of domestic violence ................... 156
Table 48: Percentage of men reporting perpetration of domestic violence ........................ 157
Table 49: Perpetration of domestic violence against individuals who reported domestic
violence and lived with their parents (in percentages) ......................................................... 158
Table 50: Responses to first social violence scenario (in percentages) ................................ 161
Table 51: Responses to second social violence scenario (in percentages) ........................... 163
Table 52: Responses to first physical violence scenario (in percentages) ........................... 167
6
Table 53: Responses to second physical violence scenario (in percentages) ...................... 169
Table 54: Percentage of respondents agreeing with wife-beating if the wife… ................. 172
Table 55: Percentage of respondents who think that a man has a good reason to hit his
wife if: ........................................................................................................................................... 173
Table 56: Percentage of respondents stating that wife-beating is acceptable if the
woman… ...................................................................................................................................... 175
Table 57: Views about statement ‘To bring up, raise or educate a child properly, the child
sometimes needs to be physically punished’ (in percentages) ............................................. 178
Table 58: Views about statement ‘If a woman wears revealing clothes, it is okay for men
to make some sexual comments as she walks in her neighbourhood’ (in percentages) .. 180
Table 59: Views about statement ‘If a woman wears revealing clothes, it is her fault if she
is raped’ (in percentages) ........................................................................................................... 181
Table 60: Regions significantly associated with increased experience of domestic violence
(by type of domestic violence and sex) .................................................................................... 197
Table 61: Relationship between experience of domestic violence and physical health
(women) ........................................................................................................................................ 204
Table 62: Relationship between experience of domestic violence and physical health
(men) ............................................................................................................................................. 205
Table 63: Percentage of respondents who have experienced domestic violence reporting
serious mental illness (by type of domestic violence and sex) ............................................. 206
Table 64: Effect of domestic violence on men and women’s mental health ....................... 207
Table 65: Percentage of respondents who experienced domestic violence who reported
seeking help (by sex) ................................................................................................................... 212
Table 66: Reasons for approaching external help after exposure to any form of domestic
violence by sex (in percentages) ................................................................................................ 212
Table 67: Reasons for not seeking external help after exposure to any form of domestic
violence by sex (in percentages) ................................................................................................ 213
Table 68: First choice of help (in percentages) ........................................................................ 214
Table 69: Respondents’ knowledge about public services (in percentages) ....................... 221
Table 70: Satisfaction levels with public services (number of respondents) ...................... 223
7
Foreword
Domestic violence, in the form of physical or sexual violence, is experienced by 35 per cent
of women globally during their lifetime. Emotional abuse and controlling behaviour is
experienced by 75 per cent of women across the world. It is also acknowledged that men
experience domestic violence. Nationally representative statistics on the incidence of
domestic violence are, however, hard to come by.
This study represents a comprehensive attempt to estimate the incidence of domestic
violence in Ghana, and analyse the attitudes that sustain domestic violence, its
determinants and its consequences. The study estimates that 28 per cent of women and 20
per cent of men experienced domestic violence in 2015.
Domestic violence causes immediate devastating consequences to those affected: physical
injuries, mental health problems and poor well-being, to name but a few. But it also has
long-term, far-reaching effects, including persistent inequalities between men and women,
which limit women and girls’ abilities to fulfil their potential. Those who experience
domestic violence have longer-term poor physical and mental health, higher risk of
contracting HIV and sexually transmitted diseases, and restricted choices in terms of
accessing education and jobs. It is also likely that they will have lower levels of education,
income and productivity. These consequences are not only life-changing for the women
and men concerned, but also for their families, communities and wider societies, as gender
inequalities become entrenched.
This innovative mixed-methods study, commissioned by the Ministry of Gender, Children
and Social Protection in Ghana and the UK Department for International Development,
and funded by UK aid, provides an in-depth understanding of the incidence, attitudes,
causes and consequences of domestic violence in Ghana, as well as investigating the
effectiveness of existing institutional support offered to women and men.
Ghana has, in many ways, led the way with its pioneering legislation on domestic violence,
which includes various forms of economic abuse, acknowledges that perpetrators and
survivors do not have to be married or related by blood ties, and takes a broader
perspective on access to justice than other countries, allowing for mediation by alternative
dispute resolution methods.
The Domestic Violence Act 732, adopted by Parliament in 2007, outlines a comprehensive
legal framework for the prevention of and protection against domestic violence and
criminalises various acts of physical and sexual violence, economic and psychological
abuse, and intimidation in domestic relations.
Despite this innovative legislative work, the results of this study show that there is more
work to be done to ensure that the laws are enacted and understood by the broader
population. The findings of this work will be used to formulate ongoing policies and
8
programmes to ensure that, moving forward, fewer and fewer people will be affected by
domestic violence in Ghana.
Hon. Nana Oye Lithur
Minister for Gender, Children and Social Protection
9
Acknowledgements
This report discusses the main findings of a comprehensive study of the incidence,
determinants and consequences of domestic violence, attitudes towards domestic violence
and the effectiveness of services that support women and men who experience domestic
violence in Ghana. The study was commissioned by the Domestic Violence Secretariat of
the Ministry of Gender, Children and Social Protection (MoGCSP) of the Government of
Ghana and the UK Department for International Development (DFID), and funded by UK
aid.
The study was led by the Institute of Development Studies (IDS) in the United Kingdom
(UK) and carried out in cooperation with the Ministry of Gender, Children and Social
Protection, the Ghana Statistical Service (GSS) and researchers at the Institute of Statistical,
Social and Economic Research (ISSER) in Accra.
The research team is very grateful to Professor Akosua Adomako Ampofo (Institute of
African Studies at the University of Ghana), who reviewed the quantitative and qualitative
instruments and participated in the quantitative pilot training exercises; to Mrs Dorcas
Coker-Appiah for providing invaluable guidance during training to the enumerating team
about domestic violence and ways of relating to potential victims of domestic violence
during the survey; and to Mr Adolf Awuku Bekoe (Coordinator, DV Coalition) for his help
in coordinating referral support provided by different organisations to the study
participants who requested it across the regions covered by the study. Mr Adolf Awuku
Bekoe also provided invaluable psychological support to the fieldwork team when needed
via a telephone helpline set up for the study.
At the MoGCSP, the study was supported by the Honourable Minister Nana Oye Lithur
and her team. The Honourable Minister chaired the study’s Steering Committee, which
provided feedback on the study’s outputs at key stages, facilitated contacts with key
stakeholders beyond the Steering Committee representatives, and supported the outreach
and uptake of the study.
The Steering Committee was comprised of Mr Kwesi Armo-Himbson (Chief Director,
MoGCSP), Mrs Efua A. Anyanful (MoGCSP), Ms Fauzia Issaka (DFID Ghana), Ms Lynne
Henderson (DFID Ghana), Mrs Patience Opoku (MoGCSP), Mrs Abena Annobea Asare
(MoGCSP), Mr Kwabena Twumasi (Human Trafficking Secretariat), Mrs Lilian Ayete-
Nyampong (Commission on Human Rights and Administrative Justice – CHRAJ), ACP
Habiba Twumasi-Sarpong (Domestic Violence Victim Support Unit – DOVVSU), Mr Adolf
Awuku Bekoe (DV Coalition), Mrs Ruth Aba Grant (Network for Women’s Rights –
NETRIGHT), Professor Felix A. Asante (ISSER), Mr Daniel Degbotse (Ministry of Health)
and Mr Clarke Noyoru (MoGCSP). The research team is grateful to Mrs Victoria Natsu for
the logistical support she provided to meetings and interactions with the Steering
Committee throughout the study.
10
The study benefited from excellent input, discussions and support from DFID Ghana. Ms
Lynne Henderson provided support to the design of the study, participated in several
discussion and reviewed drafts of the report. Mr Archie Laing provided invaluable
logistical support to the team throughout the study. Ms Fauzia Issaka engaged with the
research team at all stages of the study, provided comments on several drafts of the
fieldwork instruments, participated in numerous research discussions, helped the team to
identify and liaise with key stakeholders and reviewed in detail multiple drafts of the
report. The depth and quality of her feedback has helped substantially to improve the
report. The research team is extremely grateful for her unfaltering support and
encouragement during the entire study.
At IDS, the research team benefited from fantastic research assistance by Marco Carreras
and Eric Casper. Deborah West and Veronica Moore managed the contractual and financial
aspects of the study. Deborah West also provided invaluable assistance with
communication activities and editorial support. The study also benefited from excellent
comments and support from Dr Jerker Edstrom, Professor John Gaventa, Professor Melissa
Leach and, especially, Dr Mariz Tadros, who read and commented on the entire initial draft
of the report.
The study’s qualitative and quantitative instruments and various drafts of the report were
reviewed by four members of the study’s Advisory Board: Mrs Dorcas Coker-Appiah
(Executive Director of the Gender Centre, Accra), Dr Akosua K. Darkwah (Director, Centre
for Gender Studies and Advocacy, University of Ghana), Dr Nata Duvvury (Co-Director
of the Global Women’s Studies Programme, National University of Ireland, Galway) and
Dr Katherine Brickell (Royal Holloway University of London). The research team is very
grateful for their contribution, support and invaluable and constructive comments at all
stages of the study.
The research team would also like to thank key stakeholders from the Gender Centre, DV
Coalition, DV Secretariat, Ark Foundation, International Federation of Women Lawyers
(FIDA), Leadership and Advocacy for Women in Africa (LAWA) and other organisations,
in particular Ms Dinah Adiko (consultant, DV Secretariat, MoGCSP), Ms Juliana
Amponsah (former Acting Executive Secretary, DV Secretariat, MoGCSP), Superintendent
Laurencia Akorli (Deputy Coordinating Director of DOVVSU), Mrs Barbara Ayesu
(Coordinator, LAWA (Ghana) Alumnae Inc.), Dr James Clayman (Medical Superintendent
and Obstetrician Gynaecologist at the Ga West Municipal Hospital), Mrs Dorcas Coker-
Appiah (Executive Director, Gender Center), Mrs Esther Darko-Mensah (Gender Centre),
Mrs Angela Dwamena-Aboagye (Executive Director, Ark Foundation), Dr Elvis Sadongo
(Head of Social Services Department at Ghana Police Hospital) and Mrs Jane Quaye
(Executive Director, FIDA-Ghana) for their warm welcome, encouragement and invaluable
support.
The research team is thankful for the outstanding work of the fieldwork teams at the Ghana
Statistical Service and the Institute of Statistical, Social and Economic Research.
11
Most importantly, the research team is extremely grateful to all the women and men who
took time from their daily activities to talk about their experiences – some quite painful
and distressing. This study would have never been possible without them.
12
Research team
Core research team: Ms Alia Aghajanian (IDS), Mr Anthony Amuzu (GSS), Dr Elizabeth
Asante (ISSER), Professor Patricia Justino (IDS) (Team Leader), Dr Marinella Leone (IDS),
Dr Elizabeth Mills (IDS), Ms Sheila Minkah-Premo (Apex Lawconsult), Ms Rebecca
Mitchell (IDS), Dr Catherine Müller (IDS), Dr Philomena Nyarko (Government Statistician,
GSS), Dr Pauline Oosterhoff (IDS) and Dr Jean-Pierre Tranchant (IDS).
Fieldwork team at the Institute of Statistical, Social and Economic Research: Mohammed
Ben Abdallah, Adom K.M. Acheampong (Supervisor), Otuo Acheampong, James Allotey,
Aboagye Amankona, Dinah Badmus, Daniel Danso, George Yaw Dugbatey, Daniel Gyasi,
Samuel Gyasi-Adonteng (Supervisor), Michael Johnson, Randy Korley, Eric Obeng, Joseph
K. Okrah and Emmanuel Nuetey Siakwa (Supervisor).
Fieldwork team at the Ghana Statistical Service: Dr Philomena Nyarko (Government
Statistician/Project Director), Mr Baah Wadieh (Deputy Government Statistician/Deputy
Project Director), Mr Anthony Amuzu (Acting Deputy Government Statistician/Survey
Director), Mr Peter Takyi Peprah (Project Coordinator), Mr Emmanuel George Ossei
(Trainer/Field Monitor), Mr Emmanuel Boateng (Trainer/Field Monitor), Mr Godwin Odei
Gyebi (Trainer/Field Monitor), Mr K.B. Danso-Manu (Field Monitor), Mrs Araba Forson
(Field Monitor), Mr Owusu Kagya (Field Monitor), Mrs Betty-Love Hermans Cobbinah
(Internal Auditor), Mr Robert Kwami (Director of Finance), Mr Kwadwo Asante-Mensah
(Director of Administration), Mrs Abena Asamoabea Osei-Akoto (Head, Data Processing),
Mrs Lucy Twumwaa Afriyie (Head, Sampling Design), Ms Hannah Konadu Frimpong
(Project Secretary), Mrs Rosalind Quartey (Head, Geographic Information System) and Ms
Elizabeth Arhin Donkor (Project Accountant).
The team of regional statisticians included Isaac Addae, Salifu Amadu, Chris Amewu,
Kobina Abeka Ansah, Chris Assem, Samuel Brefo Darkwa, Sixtus Dery, Festus Manuh,
Amatus S. Nobabuma and Ernest Nyarku. The team of field supervisors included Barnas
Abugre, Michael Opoku Acheampong, Jane Acolatse, Samuel Owusu Agyemang,
Dieudonnee Ankamah, Daniel Ansah, Emmanuel Atanga, Solomon Owusu Bempah,
Victor Owusu Boateng, Patrick Kwame Djangba, Felix Geli, William Ofosu, Stephen Tabi,
Joseph, Asiedu Tenkorang and Richard Boakye Yiadom.
The team of interviewers included Daniel Amos Abanyie, Nana Afua Acheampong, Mavis
Adebona, Dennis Addo, Judith Addo, John Agbadza, Michael Agyarkwa, Frank Agyei,
Habib Alhassan, Sandra Amartey, Anastasia Ammisah, Esther Adwoa Amoah, Jolene
Mantey Amoo, Caroline Anipah, Anthony Ankomah, Umul Klusum Arimiyaw, Konlan
Alhasssan Awal, Gloria Bekoe, Vida Akuamoah Boateng, Gladys Ataa Dabison, Mabel
Diameh-Bentil, Getrude Elleammoh, Afu Anita Elorm, William Eyimah, Bright Gyamfi,
Sulemana Idriss, Felicity Kuwornu, Emmanuel Kyei, Evelyn Kyeremateng, Mark Ato
Larbi, Mary Larbi, Zainab Ibrahim Maltiti, Happy Krah Mawuse, Richard Mills, Linda
Ntiamoah, Israel Kofi Menash Nyarko, Pricilla Opoku, Seth Opoku, Mercy Naa Quarshie,
13
Phylicia Saforo, Rabiatu Seidu, Edward Tebiesinaa, Linda Ama Tenkorang, Isaac Kwame
Yeboah and Benjamin Kusi Yentumi.
The field teams were supported by a team of drivers: Ebenezer Abrokwah, Emmanuel
Acquah, Isaac Addae, Joseph Afedzie, Elvis Agyei, David Agyemang, William
Akamaboro, Joseph Ampomah, George Ansah, Eugene Aniapam, Peter Ayensu, Samuel
Ayikwei, Emmanuel Chartey, Samuel Dam, Ebenezer Dinsey, Kwesi Donkor, Tweneboah
Gyesi, Jacob Jabotir, George Karikari, Simon K. Kasakwa, Wisdom Kuagbelah, James Nii
Laryea, Victor Mensah, Gershon Nornyibey, Emmanuel Oduro, Isaac Opare, Felix Osei,
Stephen Otoo and Stephen Sanbonaa.
14
Acronyms and abbreviations
ACHPR African Charter on Human and Peoples’ Rights
CEDAW UN Convention on the Elimination of All Forms of Discrimination
against Women
CHRAJ Commission on Human Rights and Administrative Justice
CSO Civil Society Organisation
DE Design effect
DEVAW Declaration on the Elimination of Violence against Women
DFID UK Department for International Development
DHS Demographic and Health Survey
DOVVSU Domestic Violence Victim Support Unit
DV Domestic violence
DV Secretariat Domestic Violence Secretariat
EA Enumeration area
FGD Focus Group Discussion
FGM Female Genital Mutilation
FIDA International Federation of Women Lawyers
Gender Centre Gender Studies and Human Rights Documentation Centre
GFLHS Ghana Family Life and Health Survey
GSS Ghana Statistical Service
GVSSN Gender Violence Survivors’ Support Network
HIV Human Immunodeficiency Virus
IDS Institute of Development Studies
IPV Intimate Partner Violence
JHS Junior High School
JSS Junior Secondary School
LAWA Leadership and Advocacy for Women in Africa
MICS Multiple Indicator Cluster Survey
MoGCSP Ministry of Gender, Children and Social Protection
NETRIGHT Network for Women’s Rights
NGO Non-governmental Organisation
NPPOA National Policy and Plan of Action
SDGs Sustainable Development Goals
SMI Serious mental illness
UK United Kingdom
UN United Nations
UNICEF United Nations Children’s Fund
USA United States of America
VAWG Violence against women and girls
WAJU Women and Juvenile Unit of the Ghana Police
WHO World Health Organization
WISE Women’s Initiative for Self-Empowerment
15
Executive summary
Domestic violence is a violation of basic human rights and a global policy concern.
Following several years of advocacy efforts by civil society and women’s rights
organisations, the Government of Ghana enacted the DV Act (act 732) in February 2007,
followed by the formulation of the National Policy and Plan of Action (NPPOA) in 2008.
The NPPOA laid out a strategy for the implementation of the Domestic Violence Act, under
the direction of the Domestic Violence Secretariat, part of the Ministry of Gender, Children
and Social Protection (MoGCSP). This study forms part of this strategy.
The overarching aim of the study is to provide an in-depth understanding of the attitudes
to and incidence, determinants and consequences of domestic violence in Ghana. The
study addresses the following questions:
1. What types and levels of violence are experienced by women, men, girls and boys
in Ghana, and who perpetrates them? What proportion of this violence is domestic
violence?
2. What are society, community and people’s attitudes towards domestic violence, and
how do attitudes and norms influence the acceptance of different forms of domestic
violence in Ghana?
3. What are the main determinants of domestic violence against women, men, girls
and boys at the individual, family, community and wider societal levels?
4. What are the consequences of domestic violence in Ghana for people’s physical and
mental health and their daily lives?
5. What institutional structures support victims of domestic violence, and what
proportion of victims are able to effectively access available services?
6. How can policies and interventions be (re)designed to better support victims of
domestic violence and prevent domestic violence in Ghana?
Conceptual approach. The conceptual framework adopted in the study – the critical social
ecological model – uses an explicit political economy lens to analyse the complex interplay
between the determinants and consequences of domestic violence at multiple levels of
analysis: the individual, the family, the community and the wider social and political
system. This conceptual framework formed the basis for the collection of new empirical
evidence across Ghana on: (i) the incidence and types of domestic violence; (ii) the attitudes
and norms that drive domestic violence; (iii) the determinants of domestic violence; (iv)
the consequences of domestic violence; and (v) the use of and access to victims’ support
services and institutions.
Conforming to international definitions and to definitions included in the 2007 Domestic
Violence Act (Act 732), the study analysed the incidence of the following forms of violence:
Social violence: defined as acts of controlling behaviour, such as preventing someone
from seeing friends or family; stopping someone from leaving the house; requiring
to know where someone is at all times; stalking; spreading false information, videos
16
or photos without permission; or forcing women to have an abortion.
Physical violence: defined as slapping, pushing, shoving, hitting, kicking, dragging
or throwing objects at someone; choking, strangling or burning someone; using a
weapon, hazardous chemicals or substances against someone; or kicking or pulling
someone’s external genitalia (for male respondents only).
Sexual violence: defined as acts of unwanted sexual comments or physical contact;
rape by physical force, or otherwise forced sex (for instance, by blackmail or threats);
denial of using protection during sex; a sexual partner hiding their HIV status;
sexual acts and intercourse that were performed on the basis of feeling there was no
option; or penetration with an object against someone’s will.
Psychological violence: defined as the use of insults, belittling or humiliation in private
or in front of others; threats of abandonment; being ignored or treated indifferently;
intimidations and acts aimed at scaring someone; threats of using weapons against
someone; or threats of hurting someone or someone one cares about.
Economic violence: defined as the denial of household money for expenses (chop
money) even if enough financial means are available; unsolicited taking of money;
control of belongings and spending decisions; damage to or destruction of
someone’s property; denial of the right to work; forcing someone to work against
their will; or denial of food and other basic needs.
Methodological approach. The study adopted a mixed-methods approach. This involved
the collection of primary quantitative data through a representative household-level
survey of 4,995 individuals, and the collection of qualitative in-depth data on key factors
that may explain the incidence of domestic violence in Ghana.
The main quantitative tool used in the study was a survey entitled the “Ghana Family Life
and Health Survey 2015” (GFLHS 2015). The GFLHS 2015 was conducted between April
and August 2015. Given the sensitive nature of the topic, the preparation of the survey took
six months and involved intense testing, piloting and training. This careful approach
ensured high response rates, ranging between 97.6 per cent in the Western region and 99.6
per cent in the Brong Ahafo region.
The GFLHS 2015 included one household and one individual questionnaire. The study
focused on men, women, boys and girls between the ages of 15 and 60 years. This age
group extends the usual age group included in international surveys (typically 15 to 49
years), to capture the incidence of domestic violence among older women and men.
Children below the age of 15 years old were outside the scope of the study. The GFLHS
2015 asked detailed questions about individual experiences of any of the five broad
categories of domestic violence that occurred in the 12 months prior to the survey.
The sensitivity of the research topic raised several ethical and methodological challenges
that were carefully addressed during the study. The safety of respondents and the research
team was paramount, and guided all decisions in the study. Protecting the confidentiality
of the respondents was essential to ensure their safety and data quality, and all researchers
17
were carefully selected and received specialised training and ongoing support. Field
researchers were also trained to refer individuals requesting assistance to available local
services and sources of support.
Incidence of domestic violence. The study provides an in-depth analysis of the incidence,
attitudes, determinants and consequences of domestic violence in Ghana. The main results
show that 27.7 per cent of women and 20 per cent of men in Ghana have experienced at
least one type of domestic violence in the 12 months prior to the survey.
The most common form of domestic violence reported by women in the 12 months prior
to the survey was economic violence (12.8 per cent), followed by social violence (11.6 per
cent), psychological violence (9.3 per cent), physical violence (6.0 per cent) and sexual
violence (2.5 per cent).
The incidence of domestic violence was generally lower among men than among women.
The most common form of domestic violence experienced by men was psychological
violence (7.9 per cent), followed by social violence (7.7 per cent), economic violence (7.3
per cent), physical violence (2.1 per cent) and sexual violence (1.4 per cent).
The GFLHS 2015 is not comparable to prior surveys of domestic violence in Ghana because
it has collected data on broader and more in-depth definitions of domestic violence than
other surveys. It is, however, possible to compare the incidence of physical domestic
violence experienced by women and men in the 2008 DHS and in the 2015 GFLHS. The
results show, when using comparable categories of domestic physical violence, that
domestic physical violence among women decreased from 17.2 per cent in 2008 to 10.3 per
cent in 2015. Domestic physical violence among men decreased from 12.7 per cent in 2008
to 11.2 per cent in 2015.
Attitudes towards domestic violence. Respondents did not consider some of the
definitions of domestic violence used in the study (and in Act 732) to be a form of violence.
Many of the acts mentioned in the study were viewed as accepted social norms, defence
mechanisms or ways of resolving disputes. Only physical violence between partners was
consistently perceived as violence. Psychological violence, in particular, was generally
perceived as a means of defence and a coping strategy – for women in particular – to ward
off physical violence or to claim some autonomy or agency in contexts where they may not
have the ability to exert much power.
Women were on average more likely than men to condone forms of domestic violence: 23.1
per cent of women and 13.8 per cent of men found wife-beating acceptable, while 65.3 per
cent of women and 56.2 per cent of men fully agreed that women were to blame for rape if
they wore revealing clothes. These findings suggest that domestic violence in Ghana may
persist due to harmful social norms that hold women responsible for the violence they
experience.
18
Determinants of domestic violence. Age was found to be one of the key determinants of
domestic violence in Ghana. Young women and men (15–19 years old) were substantially
more likely to have experienced domestic violence than other age groups. Women aged
15–19 years were between 1.3 and four times more likely to experience any form of
domestic violence than, for instance, women aged 30–39 years. Differences in the incidence
of domestic violence across age groups were similar for men, with the exception of sexual
violence, which was more prevalent among men aged 20–24 years.
Exposure to violence in childhood was found to be strongly related to the likelihood of an
individual being a victim or perpetrator of violence in adulthood. Women who witnessed
any form of domestic violence as children were between 1.3 and three times more likely to
report domestic violence than women who did not witness violence during their
childhood. Men who witnessed social, psychological or economic violence in their homes
as children were between 1.2 and 2.2 times more likely to have experienced these forms of
violence in adulthood than other men.
Taken together, these findings suggest that Ghana may experience vicious
intergenerational cycles of domestic violence that need to be broken if further progress is
to be made in reducing domestic violence.
In addition to age, socio-economic factors were found to be among the most important
determinants of domestic violence in Ghana, albeit in complex ways:
Low asset levels were found to be strong determinants of psychological and
economic violence among women, but high asset levels were associated with higher
levels of social and sexual violence among men.
Tensions and fights over money or property were found to be leading determinants
of all forms of domestic violence across all regions.
Women with no education were generally (between 0.5 and 2.5 times) less likely to
experience domestic violence than women with some level of education. More
educated men were around three times more likely to experience economic or
sexual violence than less educated men. This result may be due to the fact that
women and men with more education may also be more likely to report domestic
violence, as emphasised in previous studies conducted internationally and in
Ghana.
The qualitative analysis suggested that factors such as poverty, unemployment and
economic shocks were identified as important determinants of domestic violence.
Overall, the determinants of domestic violence varied considerably among types of
violence, emphasising the need for domestic violence legislation in Ghana to continue to
distinguish between different forms of violence (as in Act 732).
Consequences of domestic violence. The GFLHS 2015 and the qualitative data revealed
serious adverse consequences of domestic violence on physical health, including injuries
and illness. The effects of domestic violence on physical health were observed almost
19
exclusively among women: 43.8 per cent of women who had been physically assaulted
declared having been ill in the 30 days prior to the survey – a rate that was one third higher
than among women who had not been physically assaulted (31.2 per cent). 42.3 per cent of
women who had experienced psychological violence reported ill health, compared to 30.9
per cent of women who had not experienced this form of domestic violence.
The results also showed a strong correlation between exposure to domestic violence and
mental health, particularly among women: women who reported serious mental illness
were on average around 25 per cent more likely to have experienced at least one form of
domestic violence than women who did not report serious mental illness. The association
between serious mental illness and domestic violence was generally lower among men.
Only physical violence was associated with men’s mental health status.
Domestic violence was also found to have adverse consequences on people’s daily life in
terms of their ability to work, go to school or do domestic work, their ability to concentrate
on daily activities, levels of confidence and feelings of living in fear, although few
respondents answered this question. The effects were larger for women than for men, and
for physical violence than for other types of violence.
Use of and access to services. The study investigated in detail how victims of domestic
violence make use of and access support services and institutions across Ghana. The results
showed that just over one in every three women and men who experienced domestic
violence approached a person, formal or informal organisation for help or support. The
majority of respondents who sought help first approached family members or a friend, but
14.6 per cent of men and 9 per cent of women first contacted the police for help. Individuals
who experience domestic violence appear to only resort to help beyond friends and family
in extreme circumstances because domestic violence is still considered largely a private
family matter.
Distrust of, and poor experience with, state authorities and public services also contributed
to not seeking help, a result that calls for further efforts to strengthen the services and
institutions aimed at supporting women and men affected by domestic violence in Ghana
– as well as coordination between them. Recommendations on moving forward are
provided at the end of the report, where detailed roadmaps are outlined on how best to
address the complex intersecting factors that underlie the incidence of domestic violence
among women and men and across different, socio-economic groups and locations across
Ghana.
20
1. Background to the study
Domestic violence – defined as abuse by one person against another in an intimate
relationship including marriage, cohabitation, dating or relations within the family – is one
of the most common forms of gender-based violence in the world (Abramsky et al., 2011;
Ellsberg et al., 2008; Garcia-Moreno et al., 2006, 2013; USAID, 2006). International studies
estimate that approximately 35 per cent of women across the world have experienced
physical and/or sexual violence at some point in their lives, largely in the form of domestic
violence (García-Moreno et al., 2013; Appendix A). 1 A recent systematic review has
reported that at least one in seven homicides and over one third of all female homicides
worldwide are perpetrated by an intimate partner (Stockl et al., 2013). Physical and sexual
violence are not the only types of domestic violence perpetrated against women:
emotionally abusive acts and controlling behaviour are experienced by up to 75 per cent
of women worldwide (García-Moreno et al., 2005). Domestic violence is also associated
with persistent forms of gender inequality and adverse health and economic outcomes
among victims, including poor physical and mental health, higher risks of the human
immunodeficiency virus (HIV) and other sexually transmitted diseases, restricted
livelihood options and choices, lower human capital and lower productivity (García-
Moreno et al., 2005, 2013; Moosa, 2012).
Civil society and governments around the world have acknowledged that violence against
women and girls (VAWG), including domestic violence, is a violation of basic human
rights and a global policy concern. In recognition of this international attention,
“eliminating all forms of violence against all women and girls in the public and private
spheres” is one of the targets against which Sustainable Development Goal number 5 –
aimed at achieving gender equality and empowering all women and girls – will be
measured. To further underscore the importance of tackling VAWG, the “elimination and
prevention of all forms of violence against women and girls” was chosen as the review
theme at the 60th Session of the United Nation’s (UN) Commission on the Status of Women
in March 2016.
Violence against women within private relations was drafted into international legislation
in the 1970s, as women’s movements in the United Kingdom and the United States of
America (USA) drew attention to the extent of violence committed by intimate partners.2
Domestic violence was first recognised at the international policy level in 1993 in the
1 This analysis is based on data from 155 studies in 81 countries. For further information on domestic violence
internationally, see also the Sexual Violence Research Initiative (http://www.svri.org/index.htm), and the
DFID-funded programme on What Works to Prevent Violence: A Global Programme to Prevent Violence
Against Women and Girls (http://www.whatworks.co.za/). Appendix A includes a comprehensive table
compiled by UN Women on the incidence of various forms of domestic violence across the world according
to the latest available information. 2 The first known use of the term ‘domestic violence’ in a political setting can be traced back to 1973, to an
address in the UK Parliament: in this address, domestic violence was explained to mean ‘spouse abuse,
violence in the home’: http://hansard.millbanksystems.com/commons/1973/jul/16/battered-wives. See also
Dobash and Dobash (1977), Freeman (1977) and Shanley (1993).
21
Declaration on the Elimination of Violence against Women (DEVAW), which framed
gender-based violence as a human rights violation.3 In these international declarations,
sexual and gender-based violence were defined primarily as violence committed by men
against women and girls. DEVAW, in particular, articulated three interlinked spheres in
which sexual and gender-based violence took place: in the family (including marital rape,
sexual abuse, female genital mutilation and dowry-related violence); in the community
(including rape, sexual harassment and sex trafficking); and by the State (all forms of
violence that are condoned or perpetrated by state actors).4 This approach was informed
by feminist theories that considered violence against women to be fundamentally different
from violence against men (Africa, 2010; Haraway, 1988; Yodanis, 2004). Emergent fields
of sexuality and masculinities studies have shown that domestic violence takes place
against women, against some men and also against children (Hunnicutt, 2009; see also
Institute of Development Studies et al., 2015).
This study provides a nationally representative, in-depth analysis of the incidence, levels
of acceptance, determinants and consequences of domestic violence, and the effectiveness
of institutional support offered to women and men who experience domestic violence in
Ghana. Section 1.1 below discusses in more detail current trends and research on domestic
violence in Ghana, while Section 1.2 describes the study.
1.1. Domestic violence in Ghana
The Republic of Ghana is located along the Gulf of Guinea in the sub-region of West Africa.
It has a total land area of 238,537 square kilometres and is bordered by Ivory Coast in the
west, Burkina Faso in the north, Togo in the east and the Atlantic Ocean in the south. Ghana
is divided into ten administrative regions: Western, Central, Greater Accra, Volta, Eastern,
Ashanti, Brong Ahafo, Northern, Upper East and Upper West.
3 Prior to this, human rights law did not cover gender-based violence because it excluded the ‘private sphere’
in which many forms of gender-based violence, such as domestic violence, took place. 4 1993 UN Declaration on the Elimination of Violence against Women.
22
Ghana’s population is around 24 million, according to the 2010 Census (Ghana Statistical
Service (GSS), 2013). Almost half of the total population live in the Ashanti, Eastern and
Greater Accra regions. Upper East is the least populated region, with 2 per cent of the total
population of Ghana. Over 50 per cent of the population live in urban areas (GSS, 2013).
The Ghanaian population is made up of several ethnic groups. The Akans are the largest
group (48 per cent), followed by the Mole-Dagbani (17 per cent), Ewe (14 per cent), Ga-
Dangme (7 per cent) and others (GSS, 2013). The sex ratio, according to the 2010 Census, is
95.2 males per 100 females.
According to the Demographic and Health Survey (DHS) conducted in Ghana in 2008, 38.7
per cent of ever-married women between the ages of 15 and 49 years reported having
experienced physical, psychological or sexual violence by a husband or partner at some
point in their lives. Over a quarter (27.6 per cent) of Ghanaian males reported having
experienced physical or psychological violence by their wife or partner (GSS et al., 2009).
Considerable effort has been made in Ghana over the last three decades to reduce the
incidence of domestic violence. Some of the first studies on domestic violence in Africa
took place in the 1990s in Ghana, as well as in Tanzania, Uganda and South Africa
(Hodgson, 2002; Ofei-Aboagye, 1994; Watts, Osam and Win, 1995). These studies were
motivated by the actions of activist groups, which played an important role in the
formulation and passing of domestic violence laws in Ghana (Bowman, 2002; Kimuna and
Djamba, 2008; Schneider, 2008).
23
Ghana’s first legislation efforts against domestic violence reflected the global momentum
in pushing for women’s rights to be recognised as human rights (Cook, 2011),5 prompted
by years of advocacy from key civil society organisations (CSOs), women’s rights
organisations and international bodies. These actions resulted in the Government of Ghana
enacting a number of national laws to protect women’s rights and outlaw violence against
women and girls. These included a provision in the 1992 Constitution that prohibited
discrimination based on sex; the 1998 Criminal Code Amendment Act; 6 and legal
amendments criminalising certain harmful traditional practices, such as widowhood rites
(1984), female genital mutilation (FGM) (1995) and child abuse (1998). In February 2007,
Ghana’s Parliament voted to enact the Domestic Violence Act (Act 732). Appendix B
provides a detailed historical timeline of international and Ghanaian policies related to
domestic violence.
The enactment of the 2007 Domestic Violence Act was the culmination of a long process of
advocacy and civil action, which started with the publication of the Gender Studies and
Human Rights Documentation Centre’s (Gender Centre’s) research on violence against
women and children, by Coker-Appiah and Cusack (1999), and continued under the
leadership of the Gender Centre, the Women’s Initiative for Self-Empowerment (WISE)
and the Ark Foundation.7 Responses to domestic violence were largely initiated by non-
governmental organisations (NGOs), CSOs and networks and coalitions in Ghana that
worked in the areas of human rights and women’s rights (Amoakohene, 2004; NETRIGHT,
2012).8 Together, these institutions have been engaged in activities to change perceptions,
attitudes and harmful social norms among local populations and within public institutions.
The efforts to put the Domestic Violence Act in place in the late 1990s were accompanied
by much publicity in the media. Physical spousal abuse became a regular theme in media
reports, which resulted in the establishment of the Women and Juvenile Unit (WAJU) of
the Ghana Police in 1998, a specialised unit that handled crimes against women and
children (Amoakohene, 2004). During early 2002, an increased number of stories on a series
of murders of women by their partners were publicised, fuelling further the debate on
violence against women in Ghana and enabling CSOs to mobilise public action. The role
5 These include the Convention on the Elimination of All Forms of Discrimination against Women (CEDAW),
the African Charter on Human and Peoples’ Rights (ACHPR) and the Maputo Protocol on the Rights of
Women in Africa. 6 This Act amended the Criminal Code of 1960 by increasing the age of criminal and sexual responsibility,
including a specific offence of indecent assault and revising provisions regarding sexual offences. It also
abolished customary or ritual servitude. 7 WISE provides counselling and other support services to survivors of domestic violence and has played a
central role in increasing public awareness about domestic violence. The Ark Foundation was instrumental
in forming various gender coalitions, such as the Sister’s Keepers, the Gender Violence Survivors’ Support
Network (GVSSN), the Network for Women’s Rights (NETRIGHT) and the Violence against Women and
Children’s Practitioners Network, among others. 8 These included, among others, the Ghana branch of the International Federation of Women Lawyers (FIDA-
Ghana), which worked together with other NGOs to officially abolish trokosi (a traditional religious practice
whereby human beings, usually young virgin girls, are used as payment to the Gods) and to lobby the
government to pass the Domestic Violence Bill, among other initiatives.
24
of the media at that time was criticised for their sensational framing of events and ill-
considered language that left the impression that domestic violence was something
unusual, based on individual malfunctioning rather than being situated within wider
cultural and institutional norms and practices. But Gadzepko (2009: 266) also describes the
mass media as “a key channel of information and debate on domestic violence and
consequently [...] influential in constructing public understanding of violence against
women and children”.
Efforts by civil society actors and NGOs resulted in successful mobilisation, advocacy,
sensitisation and education about domestic violence, and the training of state officials on
domestic violence and gender issues. Their efforts achieved momentum with the coming
together of stakeholder CSOs into the National Coalition on Domestic Violence Legislation,
established in 2003, which “at various times worked closely with, and at other times
independently of, or even in conflict with, the State” (Adomako Ampofo, 2008: 1).
Finally, in 2007, Ghana’s Parliament voted to enact the Domestic Violence Act (Act 732).
The actual Bill had been before Parliament since 2003 and was the “subject of heated
debate” (Adomako Ampofo, 2008). According to Takyiwaa Manuh, the “process leading
to the passage of the law involved not only the introduction of new legislation, but also
confronting a social system that tolerates various forms of violence against women and
girls, especially in the context of gender relations and in the domestic sphere” (Manuh,
2007: 1).
Ghana’s domestic violence legislation takes a broader and, arguably, culturally sensitive
approach to access to justice, when compared with other countries. First, the 2007 Domestic
Violence Act allows for mediation by alternative dispute resolution methods.9 Second, the
definition of domestic violence used in Act 732 does not include reference to a specific sex.
Third, the Act acknowledges that perpetrators and survivors do not have to be married or
related by blood ties, and applies to live-in household staff too. Fourth, the definition of
domestic violence includes various forms of economic abuse, in addition to more
conventional definitions of sexual and physical violence. The Act also provides a working
definition of domestic violence, and outlines a comprehensive legal framework for the
prevention of and protection against domestic violence. Notably, the law criminalises
various acts of physical and sexual violence, economic and psychological abuse, and
intimidation and harassment in domestic relations. These broad definitions of domestic
violence have implications for research on the incidence, nature, causes and consequences
of domestic violence in Ghana, which will be discussed in more detail in the next section.
9 Act 732 allows for the Court, with the consent of the complainant, to refer the case for settlement by an
alternative dispute resolution method if the criminal trial of a domestic violence case is not aggravated; if the
complainant requests to have the matter settled out of court; or if the court determines that the case “can be
amicably settled”. Local mediation, however, entails its own challenges, which will be discussed in Section
4.5.4.
25
1.2. The study
The Ghana Domestic Violence Act (Act 732) was followed by the formulation of the
National Policy and Plan of Action (NPPOA) developed by the former Ministry of Women
and Children’s Affairs in 2008. The NPPOA lays out the specific roles of key stakeholders
that will effectively implement the Domestic Violence Act, under the direction of the
Domestic Violence Secretariat, part of the Ministry of Gender, Children and Social
Protection (MoGCSP). This study forms part of this strategy. The overarching aim of the
study is to provide an in-depth understanding of the attitudes to and incidence,
determinants and consequences of domestic violence in Ghana. The study addresses the
following questions:
1. What types and levels of violence are experienced by women, men, girls and boys
in Ghana, and who perpetrates them? What proportion of this violence is domestic
violence?
2. What are society, community and people’s attitudes towards domestic violence, and
how do attitudes and norms influence the acceptance of different forms of domestic
violence in Ghana?
3. What are the main determinants of domestic violence against women, men, girls
and boys at the individual, family, community and wider societal levels?
4. What are the consequences of domestic violence in Ghana for people’s physical and
mental health and their daily lives?
5. What institutional structures support victims of domestic violence, and what
proportion of victims are able to effectively access available services?
6. How can policies and interventions be (re)designed to better support victims of
domestic violence and prevent domestic violence in Ghana?
The study adopted a mixed-methods approach, which involved the collection of primary
quantitative data through a representative household-level survey of 4,995 individuals –
the Ghana Family Life and Health Survey (GFLHS) 2015 – and the collection of qualitative
in-depth data on key mechanisms that may explain trends and the distribution of domestic
violence outcomes in Ghana.
Qualitative and quantitative methods were fully integrated to enable complementarity
between the methods. The quantitative data enabled the production of statistics that are
regionally and nationally representative at the household level, as well as an analysis of
determinants and consequences of domestic violence. The qualitative data explored the
perceptions, attitudes and beliefs of women and men within their everyday lives, how they
are related to domestic violence experiences as survivors and perpetrators, and how
individual experiences of domestic violence are integrated within family, community and
wider social and political structures and relations.
The next sections present the detailed findings of this study. Section 2 discusses the main
definitions of domestic violence adopted in the study, and outlines the conceptual
26
framework of the report. Section 3 explains the qualitative and quantitative methodologies
used in the study to collect nationally representative data on domestic violence in Ghana.
Section 4 discusses in detail the main findings of the study. Section 5 summarises the key
findings and discusses them in relation to previous studies conducted internationally and
in Ghana. Section 6 concludes with proposals for future policy action to reduce and prevent
domestic violence in Ghana.
27
2. Conceptual framework
Five main sets of factors have historically been used to describe the determinants of
domestic violence: individual psychological factors, sociological factors (such as family
dysfunction), male aggression, poverty and the prevalence of cultures of violence
(Bowman and Schneider, 1998; Schneider, 2000).10 International studies, largely conducted
in the USA, have emphasised the role of individual psychology, interpersonal relationships
within families and male aggression as the main explanations for the prevalence of
domestic violence (Gosselin, 2000; Heath, 2014; Perilla, Bakeman and Norris, 1994; Pollak,
2004; Schneider, 2000; Walker, 1999). These individual- and family-centric approaches to
domestic violence have been criticised for their lack of focus on factors such as poverty and
culture – arguably more relevant in developing countries. They also tend to prioritise
response mechanisms, such as psycho- and family therapy (Pollak, 2004), crime control
and retroactive responses to cater to survivors’ needs (Kelly, 2003), which may not be
useful in settings where medical and legal recourse may be prohibitively expensive and
difficult to access (Mann, 2000; Murphy, 2002). Approaches that rely on expensive therapy
to facilitate transformation may not, in addition, account for the deeper structural
inequalities that have been shown to contribute to domestic violence (Maguigan, 2002;
Schneider, 2008).
As a response, feminist theories have attempted to analyse domestic violence as an
outcome of broader contexts characterised by unequal gender relations (Bowman, 2002).
This approach has been prolific in studies on domestic violence in Africa (Banda, 1995; also
see Adjetey, 1994; Bell and Mattis, 2000; Bograd, 1999; Butegwa, 1994; Koenig et al., 2003;
Perilla et al., 1994; Sokoloff and Dupont, 2005; Tamale, 1993; Turshen and Turshen, 2000),
including one of the first domestic violence studies in Ghana by Rosemary Ofei-Aboagye
(Ofei-Aboagye, 1994a; also see Akpalu et al., 2000). These studies have found that domestic
violence against women is largely explained by women’s subordination to, and economic
dependence on, men.
Feminist theories are closely linked to cultural approaches to domestic violence. Among
these, several studies have found that some aspects of domestic violence – for instance,
wife-battering – are considered ‘normal’ within particular cultures. As an example, a study
in Nigeria revealed that police officers tended to remind wives who reported physical
violence that their ‘Yoruba culture’ permits men to beat women (Atinmo, 1997; Atinmo,
2001). Other less direct examples of ‘culture’ as an explanation for domestic violence
include the widespread practice of bride price, and gender norms that limit women’s
10 The study was informed by a comprehensive literature review and secondary data analysis. Electronic key
word searches were initially carried out in the English language in Science Direct, Scopus and Google Scholar.
Subsequently, searches for regional and international policy documents were conducted using the United
Nations and the African Commission websites. Thereafter, an extended search was done with academic
publications by following ‘related’ or ‘cited by’ articles offered on publishing journals’ websites and Google
Scholar. Studies that could not be obtained online or as hard copies in the United Kingdom were collected
during visits to Accra.
28
access to work outside their home (Abane, 2003; Armstrong, 1998). For instance, research
in Zimbabwe has found that asking for money from male partners led to physical violence
against women, as the request was interpreted as a challenge to men’s position of power
(Armstrong, 1998). Similarly, in Ghana, Amoakohene (2004: 2378) has analysed how
women are expected “to be ‘submissive to [their] husbands’, ‘be respectful’, ‘be patient’,
and ‘be dutiful and serviceable’, as tradition demands […]”.
A related approach has argued that domestic violence is explained (at least partially) by
the shift from traditional to modern, urbanised societies, a process that may result in some
men feeling threatened and more likely to find ways of asserting their power through
violence (Bowman, 2002). Abane (2003) has found that socio-economic change in Ghana
has been accompanied by increasing economic strain within households (see also
Armstrong, 1998). As a result, some men faced increased difficulties in supporting multiple
wives and households, while women took on waged employment to subsidise the family’s
income. However, women were expected to do both household and paid work. If they
failed to meet their partner’s expectations, they could be beaten. Amoakohene (2004), too,
has found that a change from traditional gender roles, where women worked at home, to
new roles, where women took on waged employment, precipitated increases in domestic
violence. International studies have revealed similar results (Justino et al., 2012).
These approaches have resulted in important insights into the causes and consequences of
domestic violence. However, most emphasise a unidimensional perspective, either
focusing on the individual, on immediate social relations or on wider social and political
processes. Structuralist theories, in contrast, have attempted to identify how drivers and
responses to domestic violence may be shaped simultaneously by social, political,
economic and cultural structures. Historically, structural approaches were championed by
some feminists in the 1970s and 1980s who, in line with CEDAW and DEVAW, argued that
domestic violence was not a private matter between individuals in the hidden spaces of
their homes. Domestic violence should, rather, be understood as a public issue shaped by
state and community actors and institutions that privilege certain gender and sexual norms
over others (Bowman and Schneider, 1998).
Within this theoretical perspective, the social ecological model (Bronfenbrenner, 1979;
Carlson, 1984; Heise, 1998; Heise and Garcia-Moreno, 2002) offered a conceptual approach
that integrated many of the above theories into a structural model that explored domestic
violence across individual, societal, economic and political spheres. The model outlined
four key levels of interaction across public and private spaces: the individual (personal
history), the family (ecosystem), the community (microsystem) and the social and political
context (macrosystem).
Subsequent critiques led to the refinement of the original model. The first critique related
to the model presenting the four layers as impermeable and unrelated to one other.
However, studies in Brazil (Minayo, 1994), South Africa (Jewkes et al., 2002) and the USA
(Perilla et al., 1994) highlighted the extent to which the experiences and personal histories
29
of the individual were strongly entwined with the ‘ecosystem’, their workplaces and
neighbourhoods. The second critique originated from post-structuralist and post-colonial
theorists who emphasised the importance of seeing individuals and structures in a circular
dynamic, where people have constrained agency within structures that they both
reproduce and challenge (Bourdieu, 2001). These theories resonated with post-colonial
critiques, particularly by African feminists (Jones, 2011; Coughtry, 2011; Ewelukwa, 2002;
Tsikata, 2014). These argued that both individualist and social ecological approaches were
ahistorical and Western-centric. Furthermore, they failed to account for the role of violently
unequal economic and social systems – such as colonialism, apartheid and globalisation –
in entrenching intersecting forms of violence, including domestic violence, which were
linked to class, race, sex, sexuality, age, location and so forth (Africa, 2010; Cornwall et al.,
2007; Jones, 2011; McFadden, 2000).
These critiques point to the importance of a conceptual framework that considers through
an explicit political economy lens the complex interplay between the individual, the family,
the community and the wider political and social system in Ghana. This study has,
therefore, adopted a ‘critical social ecological’ approach as an amalgamation of the original
social ecological theory adapted by Carlson (1984), Heise (1998) and others to domestic
violence, and the extensions proposed by Jewkes (2002). At a general level, this integrated
approach requires an understanding of individual experiences of domestic violence within
family relations, the immediate political and economic environment, local institutions
(such as the police and health care providers) and community attitudes towards violence,
and the current and historical national policy context and political and economic
environment. These four spheres and their interactions are outlined below.
The individual. Personal circumstances help to understand the attitudes, incidence,
determinants and consequences of domestic violence. As discussed above, domestic
violence often goes unrecognised, unreported and denied due to the reluctance of
individuals to report ‘private’ matters. One important example is the extent of sexual abuse
within marital relationships (Adomako Ampofo, 2008). Other examples include human
trafficking, coerced sex work, child marriage, widow inheritance and female genital
mutilation (FGM). The international literature on domestic violence has shown that
personal histories contribute to experiencing or enacting violence. Examples include
witnessing violence as a child, having experienced abuse as a child or having had an absent
father (García-Moreno et al., 2005; WHO, 2000; UNICEF, 2002). Based on these findings,
this study was designed to explore people’s experiences with domestic violence over time,
through survey questions, focus group discussions and in-depth interviews. These
methods have been able to identify people’s personal experiences of domestic violence,
their attitudes towards domestic violence, and their family and social relationships that
may account for these experiences and attitudes – from the perspectives of both survivors
and perpetrators.
The family. The family is the immediate environment in which people who experience or
exert violence live. In this sphere, factors that contribute to domestic violence include male
30
dominance and male control over the family’s wealth. These factors may also work
conversely: if men do not hold wealth and women do, women may be subject to violence
by her partner, or even by her partner’s family (Armstrong, 1998; Geisler, 1995; Njovana
and Watts, 1996). The 2008 Ghana DHS study found that the incidence of domestic violence
was associated with a complex variety of causes within the nuclear and extended family
network. The results revealed that 36.6 per cent of women aged 15–49 years experienced
physical violence, mostly perpetrated within the family by current or previous husbands
or partners, as well as by other male and female family members or acquaintances (GSS et
al., 2009). The current study, therefore, incorporated methodologies to collect data on and
analyse domestic violence from this family perspective. To this purpose, the survey
questionnaire included several questions on the determinants of violence within the
domestic sphere, and about social norms and expectations around gender roles within
immediate and wider family circles. The qualitative instruments were designed to examine
personal experiences within families, by allowing interviewees to tell personal stories
about relevant events, and explore different elements in the story to understand what
happened and why.
The community. This level of analysis takes into account the institutions and social
structures that surround the family, such as the neighbourhood, social networks and the
workplace. Human perceptions and attitudes do not emerge in a vacuum, and
communities perform important functions in the lives of individuals and their families.
Notably, the attitudes of community members, traditional and religious leaders,
neighbours, work groups and other local social groups and networks are likely to shape
the attitudes of individuals and their families, including the acceptance or tolerance of
domestic violence. In Ghana, key community figures include community leaders and local
representatives of state agencies, such as police officers and health care workers. This
current study has paid careful attention to the analysis of how attitudes and norms towards
domestic violence are shaped by wider community-level social, economic and political
dynamics. The surveys for this study were designed to capture different types of violence
experienced within the environment surrounding an individual (notably, the workplace
and the community). The survey and qualitative instruments also included questions on
attitudes and norms about the acceptability of domestic violence, about appropriate
reactions and potential punishments, and about beliefs that may affect people’s ability to
take cases of domestic violence outside the private sphere and to involve state actors, such
as the police.
The wider social and political context. Social and political contexts shape general views
and attitudes shared across societies. The original social ecological model applied to
domestic violence paid limited attention to these wider political or economic structures.
Later work by Jewkes et al. (2002) revealed how political and social institutions engage,
reproduce and challenge gender inequalities and domestic violence, and how economic
inequalities affect people’s experiences of domestic violence. Key political and social
institutions include the police, law courts, legal services, and health and social support
organisations. This is because domestic violence may go unchecked if not accompanied by
31
integrated prevention and oversight through the State’s law enforcement bodies. Domestic
violence may also be unreported if it is socially accepted across the wider society. One
objective of the current study was to uncover how institutions set up to respond to and
prevent domestic violence in Ghana may have shaped the incidence, nature, determinants
and consequences of domestic violence across the individual, their families, their
communities and the wider society. Quantitative and qualitative instruments have
measured differences in knowledge, availability, accessibility, use and satisfaction with
different services related to domestic violence. Information was also collected about views
and attitudes across the population that may explain relations between institutional
engagement, structural inequalities, gender inequality and domestic violence. The
qualitative instruments, in particular, collected data on whether people were aware of
available services, the barriers they might have encountered when accessing them, and
suggestions to support individuals and communities to hold local and national
government accountable to the provision of services listed in the 2007 Domestic Violence
Act.
The conceptual framework highlights the importance of understanding domestic violence
both within and across the different spheres that shape the incidence, distribution and
nature of domestic violence: the individual, their immediate social relations, the
community and the wider social and political system. The study used this conceptual
framework to collect new empirical evidence across Ghana on: (i) the incidence and types
of domestic violence; (ii) individual-, household- and community-level attitudes towards
domestic violence; (iii) determinants of domestic violence; (iv) consequences of domestic
violence; and (v) the use of and access to victims’ support services and institutions.
Subsection 2.1 discusses the main definitions and concepts used in the study. Subsections
2.2, 2.3, 2.4 and 2.5 describe, respectively, the current study’s approach to the collection
and analysis of information on attitudes to and determinants and consequences of
domestic violence, and how people access and use relevant services and institutions across
Ghana.
2.1. Types and perpetration of domestic violence
The most prominent international studies on domestic violence are of a quantitative
nature, based on national demographic data and health surveys (e.g. UNICEF, 2002; WHO,
2000). These typically measure physical, sexual and psychological abuse by using a set of
questions about particular acts of violence and asking respondents to state whether they
have ever experienced them, and if so, whether this happened in the 12 months prior to
the survey. These studies also usually investigate attitudes towards physical and sexual
abuse by men and women.
Studies prior to 2008 in Ghana were of a more exploratory nature but included very useful
information about context-specific and normative understandings of what constituted
domestic violence. The first study on the nature and extent of domestic violence in Ghana,
by Rosemary Ofei-Aboagye (1994), focused on women’s understandings and responses to
32
domestic violence. The study noted, in particular, the importance of terminology: only five
out of 50 women said it was accepted in their ethnic group that husbands beat their wives;
but many more did so when the term ‘disciplining’ was used. A second study by the
Gender Centre was designed to better understand the prevalence, type, contexts,
responses, barriers to responses and recommendations about domestic violence at the
national level (Coker-Appiah and Cusack, 1999). Similar to Ofei-Aboagye’s 1994 study, the
researchers asked women to define what they understood as violence. The study revealed
that most women understood that violence was not limited to physical abuse but also
included acts of psychological mistreatment such as threats, humiliations, being prevented
from seeing family and friends and sexual harassment. A later study by Amoakohene
(2004) analysed domestic violence from a more multidimensional perspective, by
exploring women’s perceptions about their rights, responsibilities, duties and the abuses
they suffer, as well as the coping strategies they developed. Domestic violence was seen to
include both physical and psychological abuse. Sexual violence was not recognised as an
offence by either the State or the community (see also Adu-Gyamfi, 2014).
These findings, alongside the definition of domestic violence proposed in the 2007
Domestic Violence Act, have informed the typology of domestic violence used in this
study. Notably, the 2007 Domestic Violence Act defines domestic violence as: “engaging in
the following within the context of a previous or existing domestic relationship:
(a) an act under the Criminal Code 1960 (Act 29) which constitutes a threat or harm to a
person under that Act;
(b) specific acts, threats to commit, or acts likely to result in
(i) physical abuse, namely physical assault or use of physical force against another person
including the forcible confinement or detention of another person and the deprivation of
another person of access to adequate food, water, clothing, shelter, rest, or subjecting another
person to torture or other cruel, inhuman or degrading treatment or punishment;
(ii) sexual abuse, namely the forceful engagement of another person in a sexual contact
which includes sexual conduct that abuses, humiliates or degrades the other person or
otherwise violates another person's sexual integrity or a sexual contact by a person aware of
being infected with human immunodeficiency virus (HIV) or any other sexually transmitted
disease with another person without that other person being given prior information of the
infection;
(iii) economic abuse, namely the deprivation or threatened deprivation of economic or
financial resources which a person is entitled to by law, the disposition or threatened
disposition of moveable or immovable property in which another person has a material
interest and hiding or hindering the use of property or damaging or destroying property in
which another person has a material interest; and
(iv) emotional, verbal or psychological abuse namely any conduct that makes another
person feel constantly unhappy, miserable, humiliated, ridiculed, afraid, jittery or depressed
or to feel inadequate or worthless;
(c) harassment including sexual harassment and intimidation by inducing fear in another
person; and
(d) behaviour or conduct that in any way
33
(i) harms or may harm another person,
(ii) endangers the safety, health or well-being of another person,
(iii) undermines another person's privacy, integrity or security, or
(iv) detracts or is likely to detract from another person's dignity and worth as a human
being.
Following this definition, the study has collected detailed information on all categories of
physical, sexual, economic and psychological abuse. In line with prior studies conducted
in Ghana and information provided to the research team in initial scoping interviews with
key stakeholders, the study also collected information on acts of controlling behaviour
(social violence). Detailed definitions of the types of violence are provided in Section 4.
Conforming to international definitions (e.g. UNICEF, 2002; WHO, 2000), the Domestic
Violence Act, and prior studies conducted in Ghana, domestic violence is defined in this
study as violence committed against a person by an intimate partner, including co-habiting
partners, and by other family members, irrespective of whether this violence occurs within
or beyond the confines of the home.
Ghana’s Domestic Violence Act provides a broad definition of what constitutes a ‘domestic
relation’, which has been adopted in this study. ‘Domestic relation’ means “a family
relationship, a relationship akin to a family relationship or a relationship in a domestic situation
that exists or has existed between a complainant and a respondent and includes a relationship where
the complainant (a) is or has been married to the respondent; (b) lives with the respondent in a
relationship in the nature of a marriage even if they are not or were not married to each other or
could not or cannot be married to each other; (c) is engaged to the respondent, courting the
respondent or is in an actual or perceived romantic, intimate, or cordial relationship not necessarily
including a sexual relationship with the respondent; (d) and respondent are parents of a child, are
expecting a child together or are foster parents of a child; (f) and respondent are family members
related by blood ties, affinity or adoption, or would be so related if they were married either
customarily or under an enactment or were able to be married or if they were living together as
spouses although they are not married; (g) and respondent share or shared the same residence or are
cotenants; (h) is a parent, an elderly blood relation or is an elderly person who is by law a relation
of the respondent; (i) is a house help in the household of the respondent; or (j) is in a relationship
determined by the court to be a domestic relationship”.
To distinguish whether acts of violence are domestic or non-domestic in nature,
respondents who reported having experienced any one of the acts of violence above during
the 12 months prior to the survey were asked whether the perpetrator was part, or not, of
the household. If the perpetrator and survivor shared the same roof and meals (the
definition of a household adopted in this study, following the standard definition adopted
by GSS), the respondent was asked to identify the perpetrator. In cases where the
perpetrator was identified as living outside the household, respondents were asked about
their relationship to the perpetrator. Answer categories included (former) partners,
extended family members living within or outside the same community, non-related
34
community members, teachers, colleagues, friends and other relations. This choice of
questions allowed the study to identify which violence was domestic in nature, as well as
to frame experiences of violence within the wider community and societal contexts.
One innovative aspect of the 2007 Domestic Violence Act is that its definition of domestic
violence does not refer exclusively to women. Although domestic violence is generally
disproportionately directed at women, some studies – almost exclusively based on data
from the developed world – have argued that this is not always the case: women can be
equally violent, and children also ought to be considered potential victims of domestic
violence (Hunnicutt, 2009). The DHS 2008 study conducted in Ghana was one of the few
studies to date to include a domestic violence module that elicited men’s experiences. Over
a quarter (27.6 per cent) of ever-married Ghanaian men aged 15–49 years in the 2008 DHS
reported having experienced physical or psychological violence by their wife or partner at
some point during their lives (GSS et al., 2009). Using the same data, Kishor and Bradley
(2012) found that 19 per cent of married women and 10 per cent of married men had
experienced physical, psychological or sexual spousal violence, confirming that it is not
only women who are victims of intimate partner violence.11 In addition, the Domestic
Violence Victims Support Unit (DOVVSU) reported to the research team during interviews
in Ghana that it had processed several cases reported by men and boys. At the international
level, similar evidence is reported in IDS et al. (2015).12 This study has attempted to uncover
the experiences of women and girls, as well as those of men and boys. It is also important
to note that different types of domestic violence and reporting behaviour of domestic
violence vary across different age groups, regions and socio-economic characteristics
(Coker-Appiah and Cusack, 1999; García-Moreno et al., 2005; GSS et al., 2009). The study
will address these differences in detail in Section 4.
2.2. Attitudes towards domestic violence
Similar to other social attitudes, attitudes towards domestic violence depend on a range of
factors. These include individuals’ beliefs about gender norms and norms that regulate the
use of violence; how such beliefs are internalised; social norms that regulate the
distribution of power and resources within the household and immediate family; and
social norms that shape gender roles and behaviours at the community and wider societal
levels (Justino, Leone and Salardi, 2015). Multi-country studies and international
comparisons have shown that women’s attitudes towards violence vary substantially
across settings. Furthermore, women living in countries where domestic violence is more
prevalent tend to find domestic violence more acceptable than women in countries where
prevalence is lower (García-Moreno et al., 2005; Appendix A). The acceptance of wife-
11 However, the authors show that women are consistently and significantly more likely than men to
experience all forms (physical, sexual and emotional) of such violence, and that intimate partner violence is
more commonly experienced by women. It is also more severe and more likely to result in injuries than
intimate partner violence experienced by men. 12 See also the research programme on Men, Boys and Gender Equality at the Institute of Development
Studies: http://menandboys.ids.ac.uk/.
35
beating also tends to be higher among women who have experienced abuse than among
those who have not (García-Moreno et al., 2005). Taken together, these findings suggest
that women seem to learn to accept violence when they are victims themselves – a result
that will be explored further in the study.
Ofei-Aboagye’s (1994) study asked respondents about the level of beating that was
perceived to be acceptable. The findings showed that one to three slaps were considered
acceptable. However, the study also found that there was a threshold beyond which
beating ‘transcended the norm’. This happened when the victim was permanently hurt,
scarred or deformed, or was beaten in public. A later study by Takyi and Mann (2006)
found that 40.5 per cent of never-married and 33 per cent of ever-married men perceived
wife-beating to be justified. The most justifiable reason given was when they perceived
that women neglected their children. Using data from the 2008 DHS, Mann and Takyi
(2009) found that women were more likely than men to believe that spousal violence
against them was justified: 64.2 per cent of men stated that abuse would never be justified,
but only 44.6 per cent of women agreed (see also Kishor and Bradley, 2012; Tenkorang et
al., 2013). These findings raise important questions about the prevalence of social norms
and beliefs that may sustain domestic violence – an issue that will be explored in detail in
Section 4.
It appears that wife-beating may be becoming less accepted. Ghana DHS statistics for 2003
and 2008 show that, across almost all regions, the percentage of respondents condoning
wife-beating decreased. The reductions were up to 35.7 percentage points among men, and
up to 45.9 percentage points among women, with the largest changes taking place in the
Upper East and Upper West regions. These were also regions with a traditionally high
prevalence of domestic violence, and where awareness-raising campaigns have
concentrated their sensitisation efforts. This study will explore how these trends may have
continued until 2015 and what social norms and changes in social norms may explain them.
2.3. Determinants of domestic violence
The existing literature on domestic violence has found evidence for strong correlations
between the incidence of domestic violence and a wide range of individual socio-economic
factors, such as age, education and marital status (Aizer, 2011; García-Moreno et al., 2005;
Jewkes et al., 2002; La Mattina, 2012). In line with the conceptual framework outlined
above, evidence shows that these individual determinants of domestic violence interact
with factors that play out at the level of families (notably, differences in intra-household
gender dynamics), as well as wider community and societal levels – in particular, the
prevalence of patriarchal norms, perceptions of masculinity and exposure to other forms
of violence. These determinants of domestic violence, and how the study was designed to
capture them, are discussed below.
Age and intergenerational effects. Several studies have shown a close association between
the incidence of domestic violence and the age of victims and perpetrators, with, in general,
36
incidence being lower among older women over their lifetime and in the recall period of
the surveys (Aizer, 2011). For example, in urban Peru, 41 per cent of women aged 15–19
years old reported having experienced intimate partner violence, in comparison to 8 per
cent among women aged 45–49 years old (García-Moreno et al., 2005). The research team
conducted a background analysis of the Ghana DHS 2008 for this study. In contrast to
international evidence, the results did not show any particular age patterns among victims
of domestic violence. The data indicated, however, that the incidence of domestic violence
in Ghana was higher among women and men whose father beat their mother. Several other
studies have found exposure to domestic violence during childhood to be positively
associated with the probability of being a victim of domestic violence as an adult (Clark et
al., 2010; Gallegos and Gutierrez, 2011; Bobonis et al., 2013). These findings are consistent
with theories of intergenerational transmission of domestic violence (Pollak, 2004). This
study will investigate further the distribution of the incidence of domestic violence across
age and generations in Section 4.
Marital status. Qualitative research has shown that, in many parts of the world, married
men are assumed to have unconditional sexual access to their wives (Ellsberg et al., 2001).
In the context of Ghana, Adu-Gyamfi (2014) has reported how attitudes, beliefs and
practices have led to violent actions against women being dismissed due to widespread
beliefs that women give up their rights to their husbands at the point of marriage,
particularly when bride prices are paid. In the international literature, marital status is also
an important factor associated with domestic violence. Women who are separated or
divorced tend to report higher incidence of domestic violence than those who are currently
married, with violence persisting after the divorce or separation (García-Moreno et al.,
2005). The timing of marriage also appears to be an important determinant of domestic
violence, with the incidence of domestic violence lower among victims who marry later
(La Mattina, 2012). The background analysis of the 2008 DHS data conducted for this study
showed that the incidence of domestic violence in Ghana was lower among currently
married individuals (women and men) than among those who were formerly married and
separated or divorced afterwards. The role of marital status on domestic violence in Ghana
in 2015 will be explored further in Section 4.
Socio-economic status and education levels. Though domestic violence affects everyone,
existing literature indicates that the incidence of domestic violence is usually associated
with low income and wealth levels (Bobonis et al., 2013). For instance, Aizer (2011) showed
that poor women were at a higher risk of abuse in the USA. In contrast, Amoakohene (2004)
reported domestic violence in Ghana to be widespread among all socio-economic classes,
with 70 per cent of educated and employed respondents in her sample reporting some
form of abuse (predominantly physical). The study by Adinkrah (2014) showed that
homicide-suicides occurred across the whole socio-economic spectrum but were more
prevalent among individuals with lower socio-economic background. Education levels are
also typically associated with incidence of domestic violence. As argued in Jewkes et al.
(2002), “education confers on individuals social empowerment, self-confidence, and the
ability to use information and resources to one’s advantage”. Aizer (2011) found that less
37
educated women were at a higher risk of abuse in the USA, while Panda and Agarwal
(2005) showed that women who owned land in Kerala (India) were less likely to experience
forms of domestic violence. García-Moreno et al. (2005) reported that in many of the
countries they analysed the prevalence of domestic violence decreased with rising
education levels. However, in some of the countries (Brazil, Namibia, Peru, Thailand and
Tanzania) the prevalence of domestic violence was lower only among women with
education levels above secondary school. This may imply that, at higher levels of
education, women acquire a greater capacity to negotiate autonomy and control of
resources within their marriage or families. Consistent with these findings, Tenkorang et
al. (2013) have shown that women with higher levels of education in Ghana in 2008 were
less likely than women with lower levels of education to experience physical violence.
Yeboah and Batse (2009) found some level of schooling (primary education) to be
correlated with higher risks of psychological violence. Education has also been found to
explain attitudes towards domestic violence: Takyi and Mann (2006) reported that younger
men and men with little education or social status are more likely than other socio-
economic groups to believe that wife-beating is justified.
According to the 2008 Ghana DHS, the incidence of domestic violence in Ghana was lower
among individuals with no education than among those with primary or higher levels of
education. La Mattina (2012) found a similar result in Rwanda. These results could suggest
that individuals who are more educated may also be more likely to be aware of existing
laws against domestic violence and to report being victims of domestic violence. The 2008
data for Ghana showed that the percentage of individuals who condoned wife-beating was
higher among those with no education than among those with some level of education.
The role of socio-economic status, and education in particular, will be further explored in
Section 4.
Intra-household gender dynamics. Several studies have attributed changes in rates of
domestic violence to changes in gender dynamics within the household. A corollary of the
feminist theories discussed in the previous section is that women who have more
autonomy and earn independent incomes may be less likely than other women to
experience domestic violence (see Goode, 1971; Dyson and Moore, 1983). For instance,
Aizer (2010) showed that increases in women’s relative wages and in women’s labour force
participation resulted in a lower incidence of domestic violence in the USA. In contrast,
others have noted that newly economically empowered women may threaten men’s
identity, resulting in increases in domestic violence. In these cases, husbands may resort to
violence to reassert their sense of power (Heath, 2014; Menon and Johnson, 2007; Tanchen,
Witte and Long, 1991). Consistent with these predictions, studies have found that increases
in women’s labour force opportunities and their access to income and other resources have
at times been associated with increases in domestic violence. This is particularly true if
women’s initial level of bargaining power before entering the labour market was low
(Bobonis et al., 2013; Heath, 2014; Hidrobo and Fernald, 2013; Justino et al. 2012).
38
In the case of Ghana, Mann and Takyi (2009) have shown that the likelihood of an
individual tolerating domestic violence was 50 per cent higher among men who
contributed less than half of the household budget (when compared to men who
contributed to the totality of the household budget). Interestingly, when men and women
were equals in terms of decision-making, men were more likely than women to believe
that abusing women was not justifiable. The background analysis conducted in this study
using the 2008 DHS data also showed that the incidence of domestic violence was lower
when men and women made decisions jointly. The incidence of domestic violence was
higher among women who made decisions about household expenses alone. These
patterns will be explored further for the 2015 data in Section 4.
Patriarchal norms. Several theories have proposed patriarchy as an explanation for the
prevalence of domestic violence, whereby violence is a way in which men exercise control
over women (Dobash and Dobash, 1979; Eswaran and Malthotra, 2011; Martin, 1976;
Wilson and Daly, 1993, 1996; Yllo and Strauss, 1990). A number of studies have reinforced
the notion of Ghana as a patriarchal country. Ofei-Aboagye (1994), for example, described
how in proverbs, oral traditions and folklore, women were celebrated for their ‘life-giving
capacities’. The duty of husbands, on the other hand, was to secure the obedience and
fidelity of the wife. Cultural traditions, such as wife inheritance and dowry payments,
forced marriages, widowhood rites and FGM, have reinforced the belief in men’s
superiority and encouraged violence against women (Adu-Gyamfi, 2014; Amoah, 2007;
Amoakohene, 2004; Adomako Ampofo, 1993; Tenkorang et al., 2013). This is particularly
true when women are dependent on men because of limited education and training
opportunities, and restricted access to cash, credit and employment (Amoakohene, 2004;
UNICEF, 2000). Mann and Takyi (2009) found further that respondents from matrilineal
groups were less likely to condone spousal abuse than those from patrilineal backgrounds
(see also Panda and Agarwal, 2005). This study will investigate further the role of
patriarchal norms as a determinant of domestic violence through the analysis of specific
questions included in the qualitative and quantitative instruments to uncover how
normative values at the community and society levels may affect individual and household
behaviours and attitudes towards domestic violence.
Masculinities. The way men express masculinity in multiple and, sometimes,
contradictory ways has been linked to gender-based violence (Edström et al., 2014a, 2014b;
Freedman and Jacobson, 2012). The focus on masculinities since the late 1980s has been due
to a number of factors with significance for sub-Saharan African countries, including the
emergence of sexually transmitted diseases such as HIV (Barker and Ricardo, 2005; Eves,
2010; Harrison et al., 2006; Mills et al., 2009), resource-scarcity and conflict-related
migration (Campbell, 1997; Jones, 2006; Cornwall and White, 2000; Little and Jones, 2000),
and social and economic crises that trigger crises of masculinities (Richards, 1996, 2005).
The ‘crisis of masculinities’ literature, however, has been recently challenged. Studies in
Sierra Leone, South Africa and Kenya, among other countries, have emphasised the
importance of fostering positive masculinities among men to encourage them to challenge
harmful patriarchal norms and collectively address gender-based violence (Mills et al.,
39
2015a, 2015b). In Ghana, Mann and Takyi's (2009) study found that expectations of
masculinity that focused on men’s financial responsibility to provide for their households
was linked to men believing that intimate partner violence was acceptable.
The pressures placed on men, through these and other harmful constructions of
masculinity, not only reinforce harmful gender dynamics and violence but have also been
found to contribute to men’s ill health. For instance, Adinkrah (2012) investigated male
suicide behaviour in Ghana using data from the Statistics and Information Technology
Unit of the Ghana Police Service. The study found that over half of the 287 (attempted)
suicides covered in the data were motivated by an attempt to avoid the dishonour,
humiliation, public disgrace and shame associated with economic failure, disease and
physical conditions, deviant or criminal conduct and sexual impotence. In a follow-up
study, Adinkrah (2014) focused on homicide-suicides. The majority of the 50 cases
analysed between 1990 and 2009 showed a particularly severe form of intimate partner
violence: they involved husbands suspecting infidelity and men whose wives threatened
them with divorce or separation. Studies on masculinities, both in and beyond Ghana,
therefore, show the harmful effects of negative masculinities on both men and women.
Emerging studies have also revealed that, as a socially constructed phenomenon,
masculinities can be positively harnessed to transform detrimental gender dynamics (Mills
et al., 2015a, 2015b).
Exposure to other forms of violence. A smaller literature has examined how domestic
violence may be related to other forms of violence prevalent within the wider society.
Recent evidence has shown that exposure to a violent conflict or civil war at some point of
an individual’s life could be associated with rises in domestic violence. La Mattina (2012)
found that exposure to the 1994 Rwandan genocide led to an increase in domestic violence
against women who married after the genocide. Gallegos and Gutierrez (2011) found a
positive and large effect of conflict exposure in Peru on the incidence of domestic violence,
particularly when conflict exposure took place during late childhood and teenage years.
The study reported, in addition, that women who were exposed to civil conflict at a
younger age were more likely to justify wife-beating. Noe and Rieckmann (2013) showed
that women living in districts in Colombia with high conflict intensity had a higher
probability of being a victim of domestic violence than those living in districts with low
conflict intensity. These results may be due to post-traumatic stress disorder or
desensitisation caused by exposure to wider violence. This literature is only in its infancy.
This study has attempted to analyse further this interesting hypothesis through questions
in the qualitative and quantitative instruments on experiences of violence, crime and safety
in and outside the communities sampled. Results are discussed in Section 4.
2.4. Consequences of domestic violence
The persistence of domestic violence is not only a violation of basic human rights but also
restricts the achievement of gender equality across a variety of social, economic and
political outcomes. Research on the consequences of domestic violence is, however,
40
limited. The few studies that have analysed the consequences of domestic violence show
that domestic violence is typically associated with a long list of adverse consequences –
notably on physical and mental health, and on the capacity of women and men exposed to
violence to carry on with their work and other daily activities. The literature has also
reported large direct and indirect effects of domestic violence within the household on
children’s physical and emotional well-being and future life prospects.
Physical and mental health. The most widely reported consequences of domestic violence
are poor physical and mental health. Adverse effects include physical injuries, pregnancy
complications, depression and higher risk of contracting HIV and other sexually
transmitted diseases (García-Moreno et al., 2005; Moosa, 2012). García-Moreno et al. (2013)
reported that women who had experienced physical or sexual abuse by their partners were
16 per cent more likely to give birth to low-weight babies than women who did not
experience intimate partner violence. They were also almost twice as likely to experience
depression, up to 1.5 times more likely to contract HIV, and more than twice as likely to
have an abortion. García-Moreno et al. (2005) showed that domestic violence has
particularly severe consequences on mental health: women who had ever experienced
physical or sexual violence by an intimate partner reported significantly higher levels of
emotional distress,13 and were more likely to have thought about suicide or have attempted
to commit suicide. In the case of Ghana, Amoakohene (2004) reported widespread adverse
psychological and emotional effects as a result of domestic violence: women in the study
mentioned being in constant fear when their husbands came home, depressed over their
lack of freedom, and suffering from stress, tension and low self-esteem. Three other recent
studies from Ghana showed harmful effects of domestic violence on taking pregnancy to
full term and the health of babies born (Pool et al., 2014), between domestic violence and
health status (Addai and Adjei, 2014), and between domestic violence and mental health
(Sipsma et al., 2013). Adu-Gyamfi (2014) reported similar non-physical side effects of
domestic violence, such as an erosion of a person’s sense of dignity, self-esteem, confidence
and respect, particularly when injuries make it difficult for victims to go out in public for
fear of being ‘found out’.
Work and daily activities. Some studies have noted a close association between
experiences of domestic violence and restricted access to livelihood options, fewer skills
and lower productivity among victims (García-Moreno et al., 2005; Moosa, 2012). Lloyd
(1997) found that female victims of domestic violence in the USA were more likely to have
been unemployed, had multiple jobs, had lower personal incomes and were significantly
more likely to receive public assistance than women who did not experience domestic
violence. Similar results were found in Friedman and Couper (1987) and Shepard and
Pence (1988), also for the USA. In Colombia, Ribero and Sanchez (2005) showed that
domestic violence against women caused increased unemployment (by 6.4 percentage
points), a 40 per cent reduction in earnings, and adverse health outcomes. In Ghana,
Britwum and Cusack (2009) found that individuals who experienced domestic violence
13 Emotional distress was identified through symptoms such as crying easily, inability to enjoy life, fatigue,
and thoughts of suicide in the four weeks prior to the interview (García-Moreno et al., 2005).
41
took time off work as a direct effect of the violence.14 Duvvury, Minh and Carney (2012)
showed that domestic violence also affects men, with men losing about 6.5 days of work
per incident of violence. Eswaran and Malthotra (2011) found a negative effect of domestic
violence on women’s autonomy, while Bloch and Rao (2002) showed that domestic
violence was a means used by perpetrators (males in their study) in India to ensure the
maintenance of their bargaining power within the household.
Child development and behaviour. Domestic violence within families is associated with
direct and indirect effects on children. The most prominent consequences include lower
birth weight, lower IQ scores and increased emotional and behavioural problems (Aizer,
2011; Koenen et al., 2003; Sternberg et al., 1993; Wolfe et al., 2003). Carrell and Hoekstra
(2010) have also shown that children who suffered from social and emotional problems
related to exposure to domestic violence also experienced lower academic achievement.
To capture the consequences of domestic violence for individuals, their families and their
communities, the quantitative survey conducted in this study included questions about the
immediate consequences of domestic violence, particularly physical and sexual violence.
Questions elicited information about whether and how violence influenced men and
women’s ability to work, attend school or provide childcare, as well as about health and
psychological consequences, including fear and confidence. This information was
complemented by a number of interviews and focus group discussions designed to
understand the consequences of domestic violence at the level of the individual, the
household and the community. Specific questions addressed social relations, the well-
being of children, and levels of cohesion within households and communities, including
tensions that may result from broken marriages and the escalation of conflicts beyond the
immediate relationship between victim and perpetrator.
2.5. Use of and access to victims’ support services and institutions
The final research question addressed by this study asks how men and women affected by
domestic violence use and access support services and institutions. Levels of domestic
violence are typically under-reported across the world. García-Moreno et al. (2005) asked
about the use of formal services across several countries (health services, legal advice and
shelters), and whether victims of domestic violence asked for support from the police,
women’s NGOs, local leaders or religious leaders. The majority of physically abused
women reported that they had never approached any formal organisation or service. When
asked about the reasons for (not) seeking help, most respondents mentioned a general
acceptance of these behaviours as normal. Others feared consequences such as further
violence, losing their children or bringing shame to their family. Seeking help – a limited
occurrence in the sample – was usually associated with the severity of the violence, the
14 This study interviewed 1,588 women aged over 19 years and 481 girls aged 13–18 years. Nearly a third
reported having experienced physical violence, a quarter had experienced psychological violence, and one
fifth had experienced sexual violence at least once in their lives.
42
potential negative impact on children and encouragement from friends and family. Only
victims of very severe acts of violence left their husbands.
Similarly, Amoakohene (2004) found that victims of domestic violence in Ghana generally
did not report abuse. Support was not always forthcoming when the victims approached
relatives, the village chief or the police. The main explanation for these findings was the
widespread view that domestic violence was normal. Other barriers that have prevented
victims of domestic violence in Ghana from seeking help from outside the family included
concern for the children’s economic and emotional well-being, lack of family support, lack
of information or economic means to pursue legal rights and the belief that women have
to obey their husbands (Coker-Appiah and Cusack, 1999; Ofei-Aboagye, 1994).
Several institutions have been set up to respond to domestic violence in Ghana but are
constrained by capacity and other logistical challenges (Mitchell, 2011). Challenges include
the lack of initial and continuing training of staff, the slow pace of the courts and the lack
of specific support services, such as counselling, emergency shelters, legal advice and
medical services. Adu-Gyamfi (2014) analysed the process of prosecuting cases of domestic
violence after the implementation of the 2007 Domestic Violence Act in the Mampong
municipality of Ghana. The study found that the number of abuse reports increased
significantly after the implementation of the Act but was not accompanied by increases in
the number of prosecutions or convictions. According to the DOVVSU 2011 annual report,
12,706 cases of domestic violence were filed in 2010. Among these, only 954 cases were sent
to court, resulting in only 118 convictions.
One of the objectives of this study was to analyse how resources and services were used in
Ghana to support victims of domestic violence, and how and when victims accessed them.
There appears to be a consensus among service users about a lack of resources and poor
implementation of relevant policies. The perceptions and experiences of authorities who
work at these specialised services are less clear. This study designed a series of structured
observations of these services (discussed in more detail in Section 3), to better understand
the opportunities and barriers for domestic violence survivors to seek and receive support.
To operationalise this framework, the study interviewed several individuals at different
levels, including opinion leaders, activists, legal experts, community leaders (religious and
political leaders, including traditional authorities), legal practitioners and journalists.
The evidence above also suggests that under-reporting of domestic violence may be due
to social normative pressures. This evidence connects to a growing body of feminist
criminology literature that challenges the focus on the criminal justice system as the main
avenue for punishing perpetrators of domestic violence (Buzawa and Buzawa, 2003; Rowe,
2015). These studies show that the apprehension and punishment of perpetrators through
penal and criminal systems might offer short-term protection for women but does not
effectively challenge patterns of violence at a societal or systemic level (Belknap, 2014;
Hoyle, 1998; Sherman et al., 1992). Recent studies show, too, that victims of domestic
violence, especially intimate partner violence, might forgo accessing the criminal justice
43
system altogether because their adjudication wishes (such as anonymity) are not met (Artz,
2014; Cerulli et al., 2015). Further, stigmatisation of those who report violence has also been
found to discourage victims of domestic violence from accessing resources available
through the criminal justice system (Sulak et al., 2014).
The study will explore these factors, alongside other potential explanations such as people
preferring other forms of intervention and mediation due to differing perceptions and trust
about the justice system and other government-appointed support structures (lawyers,
nurses, social workers and teachers, for instance), community leaders and the police. This
approach connects to a current debate in the development literature on whether or not
there is value in using ‘organic’ or ‘existing’ informal mechanisms to mediate individual-,
household- and community-level conflict (Lugo, 2015; Rowe, 2015). Research in Rwanda
and Uganda, for instance, has suggested that ‘organic’ responses foster greater levels of
reconciliation among perpetrators and victims than imposed criminal justice routes
(Doughty, 2014; Burgess and Campbell, 2015). This study will explore how men and
women perceive government institutions and actors (such as police officers and health care
providers). It will also explore how the police and other service providers view the help-
seeking behaviour of citizens. This analysis will make use of a number of questions
included in the formal survey about access to formal and informal services and support
structures, use of these structures and perceptions about their usefulness and effectiveness.
44
3. Methodology
Ghana is a culturally diverse country, and the incidence, causes and consequences of
domestic violence are likely to vary across regions, broader kinship and socio-economic
relations and individual characteristics. The study, therefore, adopted a mixed-methods
approach to collect information about the incidence, nature, determinants and
consequences of domestic violence. This approach involved the collection of primary
quantitative data through a representative household-level survey and qualitative in-
depth data on key mechanisms that may explain trends, patterns and the distribution of
domestic violence in Ghana. The qualitative and quantitative instruments were designed
to capture information related to the dimensions of the social ecological model – personal
histories; individual and household circumstances and relations within; individual and
community perceptions and realities of violence; access to, use of and quality of local
resources and services at local and regional levels; and experiences of violence perpetrated
by a range of possible actors.
From the outset, the aim of the research was to ensure complementarity between the
qualitative and quantitative research methods. The quantitative data enabled the
production of statistics that are regionally and nationally representative at the household
level. The qualitative data explored the perceptions, attitudes and beliefs of women and
men within their everyday lives, to understand how they are related to experiences of
domestic violence as survivors and victims and to observe how individual experiences of
domestic violence are integrated within family, community and wider social and political
structures and relations. The following sections describe the design of the quantitative
survey (subsection 3.1) and the qualitative instruments (subsection 3.2); sampling
procedures for the quantitative survey (subsection 3.3) and qualitative fieldwork
(subsection 3.4); the training of enumerators and pre-testing processes (subsection 3.5); the
implementation of fieldwork across all sites in Ghana (subsection 3.6); response rates
(subsection 3.7); the methods used in the analysis of the qualitative and quantitative data
(subsection 3.8); and the strict ethical procedures followed in the study (subsection 3.9).
3.1. Quantitative instruments
The main quantitative instrument used in the study was a survey entitled the ‘Ghana
Family Life and Health Survey 2015’ (GFLHS 2015). The survey deliberately did not
include any mention of ‘domestic violence’ in the title, and was described to communities
in generic terms as a survey of health and family life. Specific information about the survey
was provided only when the research team had selected the participant within each
household approached, and privacy was ensured. At that point, the survey was explained
in detail to the participant, who was offered the option of refusing to continue with the
interview. Few participants refused to continue (see subsection 3.3 below). This approach
ensured high rates of response because participants were not prevented from answering
the questionnaire by other household members worried about a survey on domestic
violence. The introduction of the study as a health and family life survey also provided a
45
level of safety to the respondents and interviewers, by avoiding any immediate conflicts
or tensions that may have arisen if the topic of domestic violence had been introduced
publicly at the outset, as well as preventing the possibility of adverse follow-up
consequences for the respondent.
The GFLHS 2015 included one household- and one individual-level questionnaire. The
design of the questionnaires was based on the World Health Organization (WHO) survey
for the Study on Women’s Health and Violence against Women (Garcia-Moreno et al.,
2005), DHS and Multiple Indicator Cluster Survey (MICS) domestic violence modules
applied in several countries (see Appendix A), the 1998 study on Violence against Women
and Children in Ghana (Coker-Appiah and Cusack, 1999) and modules used in surveys
implemented by research team members elsewhere. The questionnaire design also built on
the definition of domestic violence used in the 2007 Domestic Violence Act but included
an additional category of domestic violence to capture instances of controlling behaviour
(social violence).
As discussed in the previous section, the definition of domestic violence adopted in this
study is broader than definitions used in standard domestic violence surveys implemented
to date. The questionnaires used in the GFLHS 2015 included questions that address this
wider definition of domestic violence and, therefore, may not be directly comparable to
existing surveys such as the DHS 2008 and the 1999 Gender Centre study (Coker-Appiah
and Cusack, 1999). For instance, the DHS 2008 focused only on physical, sexual and
psychological violence. Therefore, the total incidence of domestic violence captured in this
study will not be directly comparable to the DHS findings. However, by adopting this
broader definition of domestic violence and capturing several acts of violence committed
within domestic relations but not captured before, this study provides for the first time a
comprehensive analysis of how women and men in Ghana experience all forms of domestic
violence, which can be used to fine-tune policy, programmes and monitor future progress.
The household-level questionnaire was designed to capture information about all
individuals in each sampled household. For the purpose of the study, the standard
definition of a household used by the GSS was followed, which includes a group of
individuals who normally live under one dwelling, and prepare and eat their meals
together. The household survey included standard socio-economic questions about age,
sex, education levels, children’s school attendance, occupation, work activities and related
payments, religion and ethnicity. The household questionnaire also collected information
on food consumption, household responses to economic shocks, asset ownership, decision-
making processes within the household, and conflicts in the community.
The individual-level questionnaire was administered to only one man or one woman in
each sampled household, in line with WHO (2001) guidelines. This questionnaire included
five modules on individual experiences of domestic violence from the perspective of both
victims and perpetrators. The modules followed the definitions and typologies of domestic
violence discussed in the previous section, and elicited information about experiences of
46
social, sexual, physical, psychological and economic violence. These modules were applied
to all respondents and woven into the questionnaire between other modules to avoid
repetition and respondent fatigue. Other additional modules included questions designed
to capture the various dynamics, determinants and consequences of domestic violence
across the four spheres of the social ecological model: safety in the community, social
capital, relationships within and outside the household and local community,
reproductive, physical and mental health, individual agency, access and use of services,
and approaches to parental and child discipline. The individual questionnaire also
contained a detailed module on attitudes towards domestic violence that made use of
attitudinal questions, scenario questions and vignettes, as used, for instance, in the DHS
domestic violence modules (García-Moreno et al., 2005; see also Thaler, 2012). The
complete individual and household questionnaires are included in the Questionnaire
Annex at the end of this report.
3.2. Qualitative instruments
The qualitative instruments included community-level focus group discussions,
interviews with community leaders and key informants, field observations and individual
in-depth interviews. The complete list of instruments used in the qualitative stage of
fieldwork is provided in the Questionnaire Annex at the end of this report.
The qualitative instruments were designed to generate data that could be analysed across
sex, age and location (rural/urban), and to triangulate key findings across the different
social ecological spheres. Focus group discussions generated data at the individual, family
and community levels and explored interactions across each of these levels. Interviews
engaged closely with community leaders and service providers to generate data at the level
of the community and the broader social and political contexts.
The focus group discussions took place among four distinct groups: (i) men between 18
and 50 years old who were married or living with a partner; (ii) women between 18 and 50
years old who were married or living with a partner; (iii) men and women between 18 and
25 years old who were unmarried; and (iv) men and women over 50 years old who were
married or unmarried.
The groups were disaggregated by sex and age to allow a better understanding of how
these factors may affect different responses to the research questions. Group sizes varied
between 8 and 12 people, and were guided by one facilitator and supported by one note-
taker and one translator. Further details about the composition of the groups are provided
in Appendix C. The discussions were conducted in the main local language of each region
and explored norms, beliefs and practices regarding domestic violence, as well as how and
when individuals affected by domestic violence seek medical and legal advice. Exact
questions are provided in the Questionnaire Annex at the end of this report.
47
The focus group discussions were able to draw out a great deal of detail on perceptions,
determinants and consequences of and responses to domestic violence at the level of
individuals, households, communities and the wider society. The questions not only
probed interactions across the model’s spheres and the different contexts in which domestic
violence takes place (such as asking about violence in public arenas such as market places,
compared to violence in private spaces such as homes), but they also explored different
types of domestic violence (controlling behaviour, economic violence, sexual violence and
so forth). Because the respondents were clustered in specific groups, it was also possible to
understand some of the factors (for instance, sex, age, location) that explain the incidence
of domestic violence, compared to others.
The qualitative instruments also included in-depth interviews with community leaders
and key informants, designed to understand experiences of violence and perceptions about
the causes and consequences of domestic violence, and ways and barriers to address it.
These interviews provided important insights into the practicalities of everyday work
related to domestic violence, particularly with respect to law enforcement and engagement
with community organisations dealing with domestic violence. By conducting interviews
with opinion leaders alongside the focus group discussions, it was possible to generate a
detailed picture of how people access state resources (such as health care services),
alongside the factors that affect the capacity of service providers to meet the needs of those
affected by domestic violence. These interviews enabled the team to also explore the
interplay of dynamics occurring at a political and economic level (for example, the failure
to provide resources to remote clinics or police stations), and the implications they have
for community workers, families and individuals who are affected by domestic violence.
Details about the nature and number of interviews are included in Appendix C.
The qualitative fieldwork also made use of observations made in the main health facilities
and police services of each selected district in each of the ten regions. The aim of these
observations was to examine first-hand the availability and use of public services, and
understand the barriers that women, men and service providers face. These observations
made it easier to understand the context and situations that people described in the in-
depth interviews and focus group discussions. Details of the observations conducted in the
study are provided in the Questionnaire Annex at the end of this report. The observations
were used in particular to complement individual interviews with key staff at the observed
institutions. These explored the barriers officials face in implementing national-level
policies, and their views about the relevance and effectiveness of policies to address
domestic violence.
In addition to the methods described above, 15 in-depth individual interviews were
conducted with women and men affected by domestic violence identified through specific
trigger questions in the quantitative survey. These interviews took place privately
immediately after the individual surveys by enumerators specifically trained on the use of
qualitative data collection methods (see subsection 3.3 below). These interviews allowed
more time with the respondents to discuss in detail the occurrence, perceptions,
48
consequences and ways of coping with domestic violence. Although at first the survey
team was concerned about potential respondent fatigue, it found instead that respondents
seemed “excited and appreciative to get someone to share their story with”.15 Detailed
questions are included in the Questionnaire Annex at the end of this report.
3.3. Sampling design: quantitative survey
The study focused on men, women, boys and girls between the ages of 15 and 60 years.
This age group extends the usual age group included in international surveys (typically 15
to 49 years), to capture the incidence of domestic violence among older women and men,
an issue brought to the attention of the research team during initial scoping interviews
with key stakeholders in Ghana. Children below the age of 15 years old were outside the
scope of the study due to ethical considerations and the fact that a focus on young children
would require a different research design from the one used to elicit information from
adults. The quantitative survey aimed to conduct interviews among representative
samples of individuals aged 15 to 60 years old at national, regional and urban/rural levels,
and by sex. Either one man or one woman was interviewed in each household.
The first stage of the sampling design involved compiling basic information about the
population of Ghana. According to the 2010 census, the population is spread unevenly
across the ten regions of the country, with each region having between 2.5 and 19.4 per
cent of the overall population (see Appendix D). The level of urbanisation also varies
significantly between regions, ranging between 16.3 and 90.5 per cent.16 Sex ratios are on
average balanced, varying only slightly across regions. This information was used to
stratify the population of Ghana into 20 strata at the regional level, and by rural and urban
areas within regions. The sample design then followed a two-stage procedure. The first
stage consisted of randomly selecting primary sampling units within the 20 strata using
the sampling frame of the 2010 census enumeration areas (EAs, also referred to as
‘clusters’). In the second stage, 15 households were randomly selected from a listing done
by the GSS team in an average of 30 EAs in each region.
When designing a representative sample, it is important to take into account the prevalence
of the outcome being measured (domestic violence, in the case of this study). According to
GSS et al. (2009), the prevalence of domestic violence in 2008 varied considerably between
regions (between 26.8 and 49.4 per cent for women, and between 14.3 and 42.3 per cent for
men). 17 To be confident that the GFLHS 2015 would capture the ‘true’ prevalence of
domestic violence, the study used a more conservative estimate of 50 per cent incidence
across both sexes.
15 Correspondence with Mr Peter Takyi Peprah (Project Coordinator), 3rd February 2016. 16 Urban localities were defined as localities with 5,000 inhabitants or more (Ghana Census, 2010). 17 ‘Domestic violence’ in the DHS 2008 is defined as physical, sexual or psychological violence by
husband/wife/partner against the respondent.
49
The sample size per region required to accurately estimate a prevalence of 50 per cent
across the population of Ghana is given by the following formula (Daniel, 1999):
n = 𝑍2(𝑝(1−𝑝))
𝑑2× 𝐷𝐸,
where n = sample size, z is the z statistic for a level of confidence, p is the prevalence, d is
the level of precision chosen, and DE represents the design effect. Originally, a statistical
confidence of 95 per cent was chosen, which is associated with a Z-score of 1.96. The
desired precision of the estimated prevalence was set at 5.5 per cent.18 The design effect
(DE) was calculated as follows:
𝐷𝐸 = 1 + 𝜌(𝑚 − 1),
where 𝜌 is the intra-regional correlation coefficient, and m the number of regions (10).
Similar to García-Moreno et al. (2005), the intra-regional correlation for the different types
of domestic violence in the Ghana DHS 2008 data is very low, usually around 1–2 per cent,
and never above 4 per cent. As a result, a conservative intra-regional coefficient of 7 per
cent was chosen. These calculations resulted in a sample size of 4,995 individuals in 331
EAs. Appendix D shows the allocation of the EAs for each region and describes the survey
weights.
3.4. Sampling design: qualitative fieldwork
The design of the qualitative fieldwork was based on information on the length of the
various instruments, travel times, time needed in each community, transcript preparation
times and time spent on the analysis of the data. To minimise travel costs and times, the
team selected one district in each of the ten regions of Ghana. Since in all ten regions the
regional capital covers the full district and there is no rural site, the district with the second
largest urban town within the region was selected. This ensured that both urban and rural
sites were covered within the same district. The second most populated urban town was
selected as the urban field site, and the least populated in that same district was selected
as the rural study site.
Overall, the team conducted 80 focus group discussions with women and men across
Ghana, 210 in-depth interviews with opinion leaders, activists and legal experts, and
observations and in-depth interviews with institutions of justice (20) and health (30). The
selection of participants for the qualitative fieldwork differed depending on the qualitative
tool used and type of respondent. For focus group discussions with community members
within urban areas, respondents from each peri-urban, lower-, middle- and upper-class
areas were selected using the help of local assembly staff to map out residential areas by
18 The precision parameter is related to the confidence interval of the estimate, with the width of the
confidence interval being 2d. We increased slightly this precision from the conventional 5 per cent to 5.5 per
cent due to budgetary constraints.
50
class. In rural areas, participants for the focus group discussions were randomly selected
using a list of households in each of the areas, obtained from GSS.
For interviews in and observations of police facilities, a representative at the district police
office was chosen, as well as staff of the DOVVSU unit at the district police office – if one
existed. If the district police office did not have a DOVVSU unit, the DOVVSU unit that
covered the area of the district police office was selected instead. There, the research team
interviewed the head of the unit and one assistant, and conducted on-site observations of
its services.
With respect to health facilities, the research team selected in each district two or three
members of staff from the district hospital’s outpatient department and emergency unit
and, if available, counsellors or psychologists working with individuals affected by
domestic violence. Many sampled districts in rural areas (which covered the least
populated rural area) did not have health facilities. Instead, staff of available community-
based health planning services were interviewed as part of the interviews with opinion
leaders. Exact details are provided in Appendix C.
In both urban and rural areas, traditional chiefs, the queen mother,19 a religious leader, a
headmaster or a teacher of a public school, a traditional healer and/or a private health
practitioner were selected to participate in the community/opinion leader interviews. In
urban areas, the person responsible for gender-related services at the district assembly was
also approached. Depending on availability, one or two representatives from each category
of those listed were chosen in each site. Certain categories of respondents, such as legal
practitioners and other professionals, were not resident in rural areas and were, therefore,
selected in urban sites only. All sub-groups of opinion leaders in each community were
found and interviewed (see Appendix C). Exceptions were health care workers in the
Upper East and Volta regions, where two (and not the planned three) respondents were
available to interview. These were replaced by available representatives of other eligible
groups of opinion leaders. Additional key informant interviews in each district targeted
legal practitioners identified by the Commission on Human Rights and Administrative
Justice (CHRAJ), activists from organisations working on domestic violence at national and
regional levels identified by the DV Coalition, media professionals (radio or TV) in each
district, and journalists at the national level.
3.5. Training and pre-testing
The implementation of all quantitative and qualitative instruments was preceded by
extensive training, and by in-depth pre-testing of all instruments.
19 Queen mothers are women who rule alongside the chiefs of a traditional area and who play a central role
in the traditional governance of that area. They typically come from a royal line of descendants within a
particular clan/family, usually the same as that of the traditional chief. In Ghana, queen mothers are more
prominent in the Akan culture, which practises matrilineal descent. Queen mothers’ traditional intermediary
roles have been used by international networks and local groups campaigning against domestic violence.
51
The training of the quantitative fieldwork team was done in two stages. The first training
event took place in early May 2015 using draft questionnaires. The main aim of this training
was to prepare the pre-testing of the survey instruments. The training consisted of six main
components: (i) introduction to the study and sensitisation about domestic violence in
Ghana; (ii) outline of the quantitative survey and how to introduce oneself and the study
to households; (iii) familiarisation with the questionnaires; (iv) introduction to the use of
tablet PCs and related software; (v) working through examples of the questionnaires using
tablets; (vi) discussion about research ethics; and (v) practice in the field, followed by a
detailed debrief session. The survey questionnaires were pre-tested after this training
exercise to assess the feasibility and structure of the questions, improve the flow of the
questionnaire and refine answer codes.
The re-drafted final questionnaires were then used as the basis for the second training
exercise in late May 2015. This session offered the team the opportunity to refine the
questions and answer codes further, and to adjust how questions were phrased.
Enumerators worked in different language groups to agree on the language and
terminology they would use when translating certain terms in the questionnaire. This was
important to ensure that all research methods used the right language and measured real
incidence levels and perceptions of domestic violence. The second training event also
included special exercises on qualitative methods involving members of the quantitative
survey team assigned to conduct the in-depth individual interviews.
The training of the qualitative fieldwork team took place in early April 2015. Eighteen
qualitative researchers affiliated with ISSER and 18 enumerators from GSS took part in this
event. Supervisors of the three teams (of five members each) were selected on the basis of
their leadership expertise in past research projects. The topics covered by this training
exercise included: (i) violence, gender and health, including background information on
domestic violence in Ghana; (ii) overview of the methods used for the data collection; (iii)
discussion about research ethics; (iv) training on the use of in-depth interviews and focus
group discussions, including selection criteria, planning and implementation; (v) theory,
practice and quality control with respect to notes and transcripts and to observational
methods; (vi) methods for data analysis; and (vii) planning fieldwork. The training event
was based on role-playing and mock interviews, in which scenarios were played out to
help identify the ‘do’s and don’ts’ of sensitive behaviour towards participants, and to test
the logic, length and practicality of the drafted instruments. The training of the qualitative
field team was also used as an opportunity to refine the instruments, which were tested in
the field during a piloting exercise immediately after the training event.
3.6. Fieldwork implementation
Fieldwork took place between April and August 2015. The GFLHS 2015 data collection
started on 5th June and lasted until 26th August. Fifteen teams of three enumerators and one
supervisor visited 331 EAs overall. The teams spent 2.5 days in each EA, devoting one day
52
to the listing of all households and 1.5 days to conducting 15 interviews. The quantitative
data collection was done using tablet PCs, which allowed for greater efficiency in terms of
time spent in each EA, and better precision of the answers entered into the questionnaire.20
The use of tablets also allowed the teams to send data regularly to a main server. Incoming
data were then observed through graphs and exported to a spreadsheet, which allowed
the team to look out for errors and inconsistencies in almost real time.
The qualitative fieldwork took place between 27th April and 26th July. The qualitative teams
travelled and collected data for two weeks and then returned for one week to Accra to
finalise their transcripts. They repeated this process four times. This structure was
designed to reduce pressure on the field team, and increase the quality of the transcripts.
The time spent in Accra also allowed the qualitative research lead to perform a second
round of quality checks.
Control systems were put in place to ensure quantitative and qualitative data quality. The
survey data were checked twice – first by the supervisors before uploading the data to the
server, and then by a data monitoring team, who listed and plotted key variables to look
for outliers and inconsistencies. Any abnormal findings were fed back to and addressed
by the field supervisors. The supervisors also visited the field teams regularly to carry out
quality control checks and provide technical and emotional support. With respect to the
qualitative fieldwork, interviewers peer-reviewed each other’s notes every day before
filing them. Transcripts were reviewed several times during the weeks spent in Accra.
Spending several days in any one area also allowed the qualitative team to identify gaps
and inconsistencies in time and to go back to the respondents if required.
3.7. Response rates
Table 1 shows the survey response rates. These rates ranged from 97.6 per cent in the
Western region to 99.6 per cent in Brong Ahafo.21 These high rates of response were due to
the precautions taken when administrating the survey. First, as discussed above,
information about the survey being about domestic violence was only disclosed after the
individual participant had been chosen and privacy had been ensured. Second, the
household questionnaire was designed so that it could be administered to heads of
households without disclosing the content of the individual questionnaire. This was done
because past experience of the GSS with household surveys in Ghana had shown that an
engagement with heads of households – usually male – would subsequently lead to higher
participation and completion rates among other household members. Third, the individual
questionnaire – containing the modules on experience and perpetration of domestic
violence – was administered to a randomly chosen individual aged 15–60 years using tablet
20 For example, data are only entered once, rather than recorded on paper and then entered electronically,
which minimises data entry error. 21 The percentage of completed interviews is the same as the household response rate, as there were no vacant
or destroyed dwellings or absent households, thanks to the listing exercise that preceded the survey
collection.
53
PCs. The tablets in turn enabled the research team to make use of pictorial responses, as in
García-Moreno et al. (2005), where respondents were able to choose a ‘X’, depicting no
experience of physical violence or sexual abuse (used for both separately), or a ‘tick’
depicting having been subject to physical or sexual abuse in the past (see the questionnaires
in the Questionnaire Annex at the end of this report). This choice was made at the end of
the physical and sexual violence module, without the enumerator observing it. These tools
were used to counteract potential levels of under-reporting of these types of violence, as
indicated in other studies.
Table 1: Response rates to household survey (in percentages)
W C GA V E A BA N UE UW Total
Completed 97.6 98.1 98.2 98.4 98.3 98.9 99.6 98.8 98.1 97.8 98.5
No one at
home
2.4 1.9 1.4 1.4 1.5 0.9 0.4 0.8 1.9 1.5 1.3
Refused 0.0 0.0 0.1 0.0 0.2 0.0 0.0 0.4 0.0 0.0 0.1
Dwelling
destroyed
0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Notes: W: Western region; C: Central region; GA: Greater Accra region; V: Volta region; E: Eastern region; A:
Ashanti region; BA: Brong Ahafo region; N: Northern region; UE: Upper East region; UW: Upper West
region.
The response rates to the individual survey were also consistently high (Tables 2 and 3).
Response rates for women ranged between 96.3 per cent in the Volta region and 100 per
cent in Upper West (Table 2). Response rates for men ranged between 98 per cent in Eastern
region and 100 per cent in the Upper East and Upper West regions (Table 3).
54
Table 2: Response rates to the individual survey: women (in percentages)
W C GA V E A BA N UE UW Total
Completed 98.9 99.6 98.2 96.3 99.1 99.3 99.0 98.9 99.0 100.0 98.8
Not available 1.1 0.4 1.0 1.6 0.6 0.0 0.7 0.4 1.0 0.0 0.6
First
respondent did
not allow
second
respondent to
answer
0.0 0.0 0.0 0.4 0.0 0.0 0.3 0.0 0.0 0.0 0.1
Second
respondent
refused
0.0 0.0 0.2 1.2 0.3 0.3 0.0 0.4 0.0 0.0 0.3
Lack of privacy 0.0 0.0 0.6 0.0 0.0 0.0 0.0 0.4 0.0 0.0 0.1
Other 0.0 0.0 0.0 0.4 0.0 0.3 0.0 0.0 0.0 0.0 0.1
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Notes: W: Western region; C: Central region; GA: Greater Accra region; V: Volta region; E: Eastern region; A:
Ashanti region; BA: Brong Ahafo region; N: Northern region; UE: Upper East region; UW: Upper West
region.
Table 3: Response rates to the individual survey: men (in percentages)
W C GA V E A BA N UE UW Total
Completed 99.0 99.3 98.4 98.4 98.0 98.6 98.1 99.2 100.0 100.0 98.7
Not available 1.0 0.0 1.3 0.5 0.5 0.3 0.6 0.4 0.0 0.0 0.6
First
respondent did
not allow
second
respondent to
answer
0.0 0.0 0.0 0.5 0.0 0.0 0.0 0.0 0.0 0.0 0.1
Second
respondent
refused
0.0 0.0 0.0 0.5 0.0 0.3 1.2 0.0 0.0 0.0 0.2
Lack of privacy 0.0 0.0 0.3 0.0 1.5 0.3 0.0 0.4 0.0 0.0 0.3
Other 0.0 0.7 0.0 0.0 0.0 0.6 0.0 0.0 0.0 0.0 0.2
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Notes: W: Western region; C: Central region; GA: Greater Accra region; V: Volta region; E: Eastern region; A:
Ashanti region; BA: Brong Ahafo region; N: Northern region; UE: Upper East region; UW: Upper West
region.
Table 4 presents the breakdown of the sample along key socio-demographic characteristics.
Just over a quarter (27 per cent) of the individuals sampled were aged 30–39 years. Only 8
per cent were aged 15–19 years. The majority of respondents (60 per cent) were self-
employed, and 17 per cent were employed. Over two thirds of the sample were married or
living with someone. One fifth of the individuals sampled were divorced, separated or
55
widowed, and one tenth had never been married. The most common education level was
middle school (37.7 per cent), followed by no education (19 per cent), secondary and
primary education (around 15 per cent each), higher education (9.5 per cent) and technical
school (3 per cent). In terms of asset quintiles,22 20 per cent of all individuals in the sample
are in the bottom quintile, 26.5 per cent in the second quintile, 16.6 per cent in the third
quintile, 21.2 per cent in the fourth quintile, and 15.8 per cent in the top quintile.
Table 4: Distribution of survey respondents by key socio-economic characteristics
(weighted)
Sample All All Women Women Men Men
number percentage Number percentage number percentage
Age group
15–19 394 7.9 191 6.4 203 10.3
20–24 682 13.7 380 12.7 301 15.2
25–29 936 18.8 556 18.6 378 19.1
30–39 1,362 27.4 886 29.6 475 24.0
40–49 856 17.2 512 17.1 343 17.3
50–60 744 15.0 463 15.5 281 14.2
Total 4,974 100.0 2,989 100.0 1,982 100.0
Employment
Self-employed 2,984 59.7 1,951 65.3 1,032 52.0
Employed 845 16.9 322 10.8 522 26.4
Not working 1,166 23.3 716 24.0 428 21.6
Total 4,995 100.0 2,989 100.0 1,982 100.0
Marital status
Never married 496 10.0 156 5.2 339 17.1
Married or living together 3,472 69.9 2,189 73.4 1,281 64.7
Divorced, separated or
widowed
998 20.1 638 21.4 359 18.1
Total 4,966 100.0 2,984 100.0 1,979 100.0
Residence
Urban 2,559 51.2 1,580 52.9 964 48.7
Rural 2,436 48.8 1,409 47.1 1,018 51.3
Total 4,995 100.0 2,989 100.0 1,982 100.0
22 Individuals in the lowest asset quintile have four assets from the list provided in the individual
questionnaire (see Questionnaire Annex). Individuals in the second quintile have between five and seven
assets. Individuals in the third quintile have between eight and nine assets. Individuals in the fourth quintile
have between 10 and 12 assets. Individuals in the highest quintile have between 13 and 21 (maximum) assets.
56
All All Women Women Men Men
number percentage Number percentage number percentage
Region
Western 506 10.1 280 9.4 226 11.4
Central 440 8.8 290 9.7 145 7.3
Greater Accra 949 19.0 590 19.7 351 17.7
Volta 454 9.1 251 8.4 200 10.1
Eastern 488 9.8 303 10.1 185 9.3
Ashanti 954 19.1 575 19.2 378 19.1
Brong Ahafo 507 10.1 331 11.1 175 8.8
Northern 411 8.2 216 7.2 192 9.7
Upper East 173 3.5 86 2.9 86 4.4
Upper West 114 2.3 67 2.2 45 2.3
Total 4,995 100.0 2,989 100.0 1,982 100.0
Education level
None 958 19.3 684 22.9 274 13.8
Primary 751 15.1 504 16.9 246 12.4
Middle/JSS/JHS* 1,871 37.7 1,167 39.1 703 35.6
Secondary 769 15.5 361 12.1 407 20.6
Technical 144 2.9 78 2.6 66 3.4
Higher 472 9.5 190 6.4 281 14.2
Total 4,964 100.0 2,984 100.0 1,977 100.0
Asset quintile
Lowest 999 20.0 583 19.5 410 20.7
Second 1,324 26.5 802 26.8 517 26.1
Third 828 16.6 487 16.3 338 17.1
Fourth 1,057 21.2 667 22.3 384 19.4
Highest 788 15.8 450 15.1 332 16.8
Total 4,995 100.0 2,989 100.0 1,982 100.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Note: Differences in totals are due to missing values in the respective questions.
* JSS = junior secondary school; JHS = junior high school.
The survey captured a good balance of male and female respondents in the household
interview. There are, however, more female respondents (60 per cent) in the individual
interviews because men were much more often away from the household than women – a
common feature of most household surveys. This did not affect the analysis or main
results, which control for sex composition.
The geographical representativeness of the sample was also close to the census
distributions. Table 4 shows the distribution of the sample across rural and urban areas:
48.8 per cent in rural areas (49.1 per cent in the 2010 census) and 51.2 per cent in urban
areas (50.9 per cent in the 2010 census). It also shows the distribution of the sample across
regions: 10.1 per cent in the Western region (9.6 per cent in the 2010 census), 8.8 per cent in
the Central region (8.9 per cent in the 2010 census), 19 per cent in Greater Accra (16.3 per
57
cent in the 2010 census), 9.1 per cent in the Volta region (8.6 per cent in the 2010 census),
9.8 per cent in the Eastern region (10.7 per cent in the 2010 census), 19.1 per cent in the
Ashanti region (19.4 per cent in the 2010 census), 10.1 per cent in the Brong Ahafo region
(9.4 per cent in the 2010 census), 8.2 per cent in the Northern region (10.1 per cent in the
2010 census), 3.5 per cent in the Upper East region (4.2 per cent in the 2010 census) and 2.3
per cent in the Upper West region (2.3 per cent in the 2010 census).
3.8. Data analysis
The quantitative survey data were weighted to account for differences in the selection
probability and in the number of completed interviews among primary sampling units (see
details in Appendix D). The quantitative data were analysed using cross-tabulations
between domestic violence outcomes and socio-economic variables such as age, sex,
region, educational level, marital status, employment status and wealth, among others.
The study also used multivariate regressions to understand in more detail the
determinants and consequences of the various forms of domestic violence.
Regression analysis account for the correlations between potential determinants or
consequences of domestic violence – something that is not possible when using only cross-
tabulations. For instance, individuals with higher levels of education also tend to be
wealthier. It is then possible that both higher levels of education and higher levels of wealth
are associated with higher incidence of domestic violence in isolation but not together. It
may be that, in fact, one of those factors is more important than the other. Alternatively,
both variables may be independent drivers of domestic violence. Multivariate regressions
help to identify the most significant determinants of domestic violence by taking into
account correlations between those determinants.
Second, multivariate regressions enable the analysis of regional differences. Domestic
violence in Ghana varies considerably across regions. This may be because regions are
different in the ways in which domestic violence occurs. However, the same result may
also be explained not by genuine cultural differences between regions but, rather, because
regions with higher levels of domestic violence are also poorer or have higher levels of less
educated populations. Multivariate regressions distinguish between these two
explanations. This information is, in turn, important for designing the most effective
policies.
Third, multivariate regressions allow the inclusion within a single framework of
determinants of violence at the individual, household and wider community or national
level – in keeping with the social ecological framework of domestic violence discussed in
Section 2. It is important to note that while multivariate regressions enable the
identification of the most significant determinants and consequences of domestic violence
in Ghana, the results suggest an association between the variables but do not indicate
causality.
58
The analysis of the qualitative data was based on over 300 transcripts. The data were
analysed using Grounded Theory (Glaser and Strauss, 1998), 23 which encourages an
iterative approach. The qualitative data were analysed using a three-fold analytical system.
First, the data were clustered by region. Two researchers each reviewed data from three
regions, and one researcher reviewed data from four regions. Data were then coded against
the main research questions. This entailed systematically combing through the transcripts
from each region and looking for evidence that related to each of the main research
questions. This first stage of analysis was collated, and circulated to other research team
members to identify any initial queries or inconsistencies. In the second stage, the research
team reviewed one another’s coding, and suggested areas that required further
substantiation or clarification. In the third stage of the process, the researchers generated a
narrative analysis for each research question, based on the empirical material. This analysis
was circulated again among the research team for validation.
Through these three stages of coding, the research team sought to ensure that the claims
made in one set of transcripts, in one region, were validated by triangulating the findings
from these transcripts with data from other regions. During the coding, the following rules
were applied to ascertain validity across a scale of representativeness. ‘Nationally
representative’ findings are those findings that are corroborated by data in all regions, in
both rural and urban sites, and in data from a range of stakeholders (in the key informant
interviews) and group compositions of the focus group discussions. ‘Regionally
representative’ findings are those findings that are supported by data in both rural and
urban sites in one region, and in data from a range of stakeholders (in the in-depth
interviews) and group compositions of the focus group discussions.
3.9. Ethical considerations
The sensitivity of the research topic raised serious ethical and methodological challenges.
Notably, the study involved the collection of data that could have potentially affected the
privacy of subjects, and may have induced some level of stress and anxiety. The study also
involved potential safety risks for interviewers and respondents. Accordingly, the study
followed closely the Ethical and Safety Guidelines developed by the WHO (2001) for
conducting research on domestic violence.
Research was conducted professionally and ethically, with strict respect for principles of
integrity, honesty, confidentiality, voluntary participation, impartiality and the avoidance
of personal risk. Adherence to these principles was overseen and monitored by the team
supervisor and field monitors, in collaboration with the IDS Research Ethics Committee.
The detailed mechanisms that were put in place to ensure compliance with the principles
included:
23 Grounded Theory is a systematic methodology that groups concepts and categories through induction
based on data collected.
59
1. The safety of respondents and the research team was paramount, and guided all decisions
in the study. To ensure the safety of both respondents and the research team, the survey
was introduced to communities and households as a study on women’s health and family
life, as discussed earlier. Interviews were conducted in private settings at convenient and
safe times for the participant, and interviewers were trained to change the subject to a less
sensitive topic if an interview was interrupted. The survey participants were informed
about this beforehand. The safety of interviewers was also ensured by planning visits to
research sites in teams, particularly during the evening and to areas known to be unsafe.
2. Protecting confidentiality was essential to ensure both safety and data quality. The
questionnaires were implemented and analysed anonymously, with the names of
individuals in each household replaced by a numeric code (identifier). Respondents were
informed about protocols for confidentiality and anonymity. No names or addresses
appeared on the questionnaires or transcripts. The survey questionnaires were linked to the
in-depth interview transcripts via the personal identifiers only. The address of individuals
willing to participate in in-depth interviews was noted when they consented to be
interviewed, and destroyed after completion of the interview. Interviewers were trained
and instructed to obtain the permission of respondents to record interviews beforehand
(when this was necessary, for the in-depth interviews and focus group discussions). Tapes
and digital recorders were locked in safe places at all times, and destroyed after
transcription.
3. All research team members were carefully selected and received specialised training and
ongoing support. The training entailed an introduction to gender, gender discrimination
and domestic violence, as discussed earlier. Training also addressed the issue of
stereotypes, biases and fears regarding domestic violence, and provided mechanisms to
overcome these. Field researchers learned and practised how to ask questions in a
supportive and non-judgemental manner, and were given opportunities to come to terms
with emotions evoked during the interview process or by the topic itself (for instance, if the
field researcher had experienced abuse her/himself). Selected field researchers were
encouraged to think about and discuss openly or in private any matters of concern with
training staff and field managers. Field researchers who faced emotional distress during
fieldwork were free to take breaks, carry out less emotionally draining tasks or withdraw
from the fieldwork altogether without judgement. Field researchers were also trained to
regard themselves as researchers and not counsellors. However, they received sufficient
and adequate information to assist those respondents who asked for help (in the form of
pamphlets that directed study participants to appropriate services and counselling).
4. The study design included actions aimed at reducing any possible distress caused to the
participants by the research. As discussed above, the language of the questionnaire and the
behaviour of field researchers was non-judgemental and supportive at all times, and
interviewers were trained on how to respond to distress caused by the interviews. They
were also encouraged and learned how to terminate interviews if necessary. All interviews
60
ended on a positive note, reminding the participants of their self-worth and of the
importance of having shared their experiences.
5. Field researchers were trained to refer individuals requesting assistance to available
local services and sources of support. In cooperation with government agencies and
services, local partners and other stakeholders, such as NGOs and women’s groups, the
research team identified and obtained consent from local formal and informal providers to
direct to them individuals in need. Pamphlets containing discreet information about these
services were given to respondents, regardless of whether they reported being affected by
domestic violence or not.
6. Findings will be properly interpreted and used to advance policy and intervention
development. The research study involved from the outset formal and informal groups
involved in addressing domestic violence, mitigating its consequences and increasing the
well-being of those exposed to domestic violence (see the composition of the Steering
Committee). The research team sought their advice and input at key stages of the study.
The study also paid close attention to the language used when interpreting results, to avoid
stigmatisation or reinforcement of stereotypes of sub-groups. Similar procedures will be
employed when disseminating the results of the study.
61
4. Empirical results This section presents the main findings of the study for each of the research questions
outlined in Section 1. This analysis brings together the qualitative data, and the descriptive
statistics and regression analysis based on the GFLHS. This joint analysis of the qualitative
and quantitative data has allowed the research team to identify points of similarity and
contrast between the data collected using different methods. Each of the subsections below
presents and discusses the main results of the study against each research question in turn.
Subsection 4.1 analyses the incidence of social, physical, sexual, psychological and
economic violence across different population groups in Ghana – independently of
whether this violence was perpetrated or not within domestic relations. Subsection 4.2
focuses on the main remit of the study – the analysis of violence against women and men
perpetrated within domestic relations (i.e. domestic violence). Subsection 4.3 examines the
profiles of perpetrators of domestic violence in Ghana. Subsection 4.4 analyses attitudes
and social norms that drive domestic violence in Ghana across individuals, families,
communities and the wider society. Subsection 4.5 discusses the main determinants of
domestic violence uncovered in the data at the individual, family, community and wider
societal levels. The consequences of domestic violence are analysed in subsection 4.6, with
a focus on physical and mental health and the impact of domestic violence on daily lives
and children. The final subsection (4.7) analyses how women and men affected by domestic
violence in Ghana access and use institutions and available public services.
4.1. Incidence of violence (domestic and non-domestic) against women and men
in Ghana
Conforming to international definitions (e.g. UNICEF, 2002; WHO, 2000), the national 2007
Domestic Violence Act (Act 732), and prior studies conducted in Ghana, the study has
analysed the incidence of the following forms of violence (both within and outside
domestic relations):
social violence – acts of controlling behaviour, such as preventing someone from
seeing friends or family of birth; stopping someone from leaving the house;
requiring to know where someone is at all times; stalking; spreading false
information, videos or photos without permission; or forcing women and girls to
have an abortion;
physical violence – slapping, pushing, shoving, hitting, kicking, dragging or
throwing objects at someone; choking, strangling or burning someone; using a
weapon, hazardous chemicals or substances against someone; or kicking or pulling
someone’s external genitalia (for male respondents only);
sexual violence – acts of unwanted sexual comments or physical contact; rape by
physical force, or otherwise forced sex (for instance, by blackmail or threats); denial
of using protection during sex; a sexual partner hiding their HIV status; sexual acts
or intercourse that were performed on the basis of feeling there was no option; or
penetration with an object against someone’s will;
62
psychological violence – insults, belittling or humiliation in private or in front of
others; threats of abandonment; being ignored or treated indifferently;
intimidations and acts aimed at scaring someone; threats of using weapons against
someone; or threats of hurting someone or someone one cares about; and
economic violence – denial of household money for expenses (chop money) even if
enough financial means are available; unsolicited taking of money; control of
belongings and spending decisions; damage to or destruction of someone’s
property; denial of the right to work; forcing someone to work against their will; or
denial of food and other basic needs.
The GFLHS 2015 asked detailed questions about individual experiences of any of these five
broad categories of violence over the individual’s lifetime, and in the 12 months prior to
the survey. This allowed the study to capture both current incidence levels of overall levels
of (domestic and non-domestic) violence and the lifetime accumulation of experiences of
overall violence among men and women. Table 5 summarises this information.
Table 5: Incidence of violence in Ghana 2015 (in percentages)
Social
violence
Physical
violence
Sexual
violence
Psychological
violence
Economic
violence
Total
Percentage of respondents who experienced violence over their lifetime
Women 41.0 42.4 30.0 48.0 28.2 71.5
Men 34.4 50.8 23.1 51.9 25.8 71.4
Percentage of respondents who experienced violence in the 12 months prior to the survey
Women 20.8 8.9 10.6 22.5 14.4 42.9
Men 18.7 10.2 9.5 28.0 12.2 43.6
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Table 5 shows that 71.5 per cent of women and 71.4 per cent of men reported having
experienced at least one form of violence (domestic and non-domestic) over their lifetime,
and 42.9 per cent of women and 43.6 per cent of men experienced at least one form of
violence in the 12 months prior to the survey. The following subsections discuss in detail
how each of these forms of violence was experienced by the respondents of the GFLHS
2015 across their characteristics and geographical location.
4.1.1. Social violence
Table 5 shows that 41.0 per cent of female respondents and 34.4 per cent of male
respondents experienced at least one type of social violence over their lifetime, committed
by perpetrators both within and outside domestic relations. Just over 20 per cent of women
and 18.7 per cent of men experienced at least one form of social violence during the 12
months prior to the survey. The survey explored further the incidence of social violence
along its different dimensions (as defined above), and across regional and socio-economic
63
characteristics of the sample. This disaggregated information is included in Tables 6–9 and
discussed in the following paragraphs.
Dimensions of social violence. Social violence experienced by both men and women over
their lifetime was mostly in the form of rumours, photos or videos. Almost a quarter (24.0
per cent) of women (Table 6) and 22.8 per cent of men (Table 8) experienced this form of
violence. The second and third most common forms of social violence over the lifetime
were, respectively, not being allowed outside their homes (experienced by 14.4 per cent of
women and 10.6 per cent of men), and being kept away from family and friends
(experienced by 14.1 per cent of women and 8.9 per cent of men). These patterns were
similar when considering the incidence of social violence in the 12 months prior to the
survey (Tables 7 and 9).
Age patterns. The incidence of social violence among men and women was strongly
associated with age, with older people being less likely to have experienced social violence.
For example, 53 per cent and 49.4 per cent of women aged 15–19 and 20–24 years,
respectively, reported having experienced social violence over their lifetime. This was in
contrast with 33.1 per cent of women aged 40–49 years and 33.7 per cent of women aged
50–60 years (Table 6). There were also large differences by age among men: almost 40 per
cent of young men aged 15–19 years and 36.6 per cent of men aged 20–24 years reported
having experienced social violence over their lifetime (Table 8). The incidence of social
violence over the lifetime of men aged 40–49 years was 29.7 per cent. It was 32.2 per cent
among men aged 50–60 years. These age patterns were similar when considering the
incidence of social violence in the 12 months prior to the survey (Tables 7 and 9).
Employment status. The incidence levels of social violence over the lifetime of women and
men were slightly higher for women not working (i.e. not in waged employment): 42.8 per
cent of currently non-working women reported having experienced social violence,
compared to 40.5 per cent of self-employed women and 40.3 per cent of employed women
(Table 6). Social violence was most prominent among employed men (37.6 per cent),
followed by men who were not working at the time of the survey (34.6 per cent) and self-
employed men (32.7 per cent) (Table 8). Tables 7 and 9 show similar patterns across
employment status of women and men for the incidence of social violence in the 12 months
prior to the survey.
Marital status. The results indicate that married or divorced, separated or widowed men
and women were more likely than those never married to have experienced social violence.
Almost 41 per cent of women who were married or living with a partner, and 41.5 per cent
of women divorced, separated or widowed, experienced social violence, in comparison to
39.9 per cent of women who were never married (Table 6). Thirty per cent of never-married
men reported having experienced at least one form of social violence over their lifetime, in
comparison to, respectively, 35.1 and 35.8 per cent of men who lived with a partner (or
were married) or men who were divorced, separated or widowed (Table 8). Women who
were never married (32.8 per cent) reported higher levels of social violence in the 12
64
months prior to the survey than women with other marital status (this was the reverse of
the pattern observed for lifetime social violence) (Table 7).
Geographical patterns. The incidence of lifetime social violence was higher among women
living in rural areas (42.3 per cent) than women living in urban areas (39.9 per cent). It was
also higher among women living in the Eastern (67.7 per cent) and Western regions (48.3
per cent) (Table 6) than for those living in other regions. The incidence of lifetime social
violence among men showed a similar pattern (Table 8). Differences across urban and rural
areas were less pronounced among women for the 12 months prior to the survey than
among women reporting lifetime experiences of social violence. Among the former,
women living in urban areas were slightly more likely than women in rural areas to have
experienced this form of violence (21.5 per cent, in comparison to 20.1 per cent in rural
areas) (Table 7). Men living in urban areas were also less likely to report having
experienced social violence over the 12 months prior to the survey than men in rural areas
(21.5 per cent, in comparison to 16.1 per cent of men living in rural areas – a reverse of the
pattern observed among men (and women) reporting social violence over their lifetime)
(Table 9). These differences may be due to recent changes in media technology (since the
main form of social violence was the spread of rumours, videos and photos), which are
more likely to be more widespread in urban areas.
Education levels and asset ownership. The incidence of lifetime social violence was lower
among women with no formal education (31.8 per cent) and among women with higher
(university) education (35.1 per cent) than among women with intermediate levels of
education (primary, middle, secondary and technical) (Table 6). Men with no education or
with technical education were less likely to report social violence (25.6 per cent and 26.2
per cent, respectively), than men with other levels of education (Table 8). The results show
no clear association between the lifetime incidence of social violence and asset ownership
for men or women. These education and wealth patterns were similar when considering
the incidence of social violence in the 12 months prior to the survey (Tables 7 and 9).
65
Table 6: Percentage of women who have experienced domestic or non-domestic social violence over their lifetime
Kept from
seeing
friends or
family
Stopped from
leaving home
Asked to
report
activities
Stalked Rumour,
photos or
videos spread
Forced to
have an
abortion
Any social
violence
Age group (years)
15–19 23.0 27.8 15.4 4.2 24.6 6.7 53.0 20–24 20.1 20.7 18.1 6.0 24.9 9.4 49.4 25–29 16.7 13.2 12.7 4.2 27.6 12.7 47.7 30–39 11.1 12.4 10.0 2.9 24.1 8.5 39.0 40–49 10.8 10.3 5.2 1.2 20.7 7.2 33.1 50–60 11.6 13.3 5.2 1.8 22.1 5.1 33.7
Employment
Self-employed 13.7 14.0 8.6 2.7 25.2 8.2 40.5 Employed 11.4 11.3 13.8 3.5 21.5 9.5 40.3 Not working 16.2 16.8 13.4 4.2 21.8 9.1 42.8
Marital status
Never married 18.8 19.9 10.1 1.6 17.6 1.5 39.9 Married or living together 13.4 14.0 10.7 2.9 23.9 8.8 40.9 Divorced/separated/widowed 15.2 14.2 9.1 4.2 26.1 9.3 41.5
Residence
Urban 13.2 12.4 11.6 3.8 22.9 8.9 39.9 Rural 15.0 16.6 8.8 2.4 25.2 8.2 42.3
66
Kept from
seeing
friends or
family
Stopped from
leaving home
Asked to
report
activities
Stalked Rumour,
photos or
videos spread
Forced to
have an
abortion
Any social
violence
Region
Western 14.9 16.0 6.0 1.1 32.3 12.7 48.3 Central 12.4 15.4 7.7 2.4 18.6 11.5 38.1 Greater Accra 8.9 6.7 9.6 3.0 17.0 6.1 29.4 Volta 10.7 10.9 9.3 3.6 37.3 4.9 47.3 Eastern 35.2 41.0 17.1 6.3 42.0 11.9 67.7 Ashanti 7.0 9.0 12.1 2.4 25.9 11.7 42.2 Brong Ahafo 17.2 10.1 13.8 4.4 16.2 9.7 40.3 Northern 19.5 19.7 4.9 2.9 13.2 0.9 31.0 Upper East 15.1 19.2 3.3 2.7 14.7 0.0 27.3 Upper West 8.0 7.5 12.6 1.1 11.1 2.3 24.6
Education level None 13.0 12.9 5.1 1.5 18.7 5.3 31.8
Primary 17.5 18.5 10.1 3.5 29.3 9.8 46.2 Middle/JSS/JHS 14.1 14.9 10.6 3.3 26.7 10.7 44.5 Secondary 13.0 12.8 16.8 4.3 19.9 7.6 42.6 Technical 12.7 11.3 11.9 3.3 28.8 13.5 43.8 Higher 11.6 10.3 14.1 5.0 18.2 3.6 35.1
Asset quintile
Lowest 15.4 16.2 6.9 2.5 24.8 6.9 41.3 Second 15.0 15.7 9.7 3.5 24.4 8.8 41.2 Middle 12.3 13.5 10.6 3.3 26.1 11.6 43.8 Fourth 14.0 13.5 12.2 3.5 23.0 9.0 39.4 Highest 12.9 11.9 12.7 2.8 21.4 6.3 39.6
Total 14.1 14.4 10.3 3.1 24.0 8.5 41.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
67
Table 7: Percentage of women who experienced domestic or non-domestic social violence in the last 12 months
Kept from
seeing
friends or
family
Stopped from
leaving home
Asked to
report
activities
Stalked Rumour,
photos or
videos spread
Forced to
have an
abortion
Any social
violence
Age group (years)
15–19 16.9 21.8 12.2 3.9 17.9 3.9 44.2 20–24 7.8 7.1 12.7 3.7 16.7 2.1 31.5 25–29 5.0 4.3 8.4 2.4 17.7 0.9 25.7 30–39 2.4 2.3 4.2 1.7 15.2 0.4 18.7 40–49 1.3 1.5 2.3 0.8 9.6 0.0 12.7 50–60 0.9 0.7 0.7 0.4 8.2 0.0 9.8
Employment
Self-employed 2.4 2.5 3.7 1.5 14.0 0.6 18.1 Employed 2.2 2.7 7.0 1.2 14.0 1.5 22.7 Not working 9.3 9.3 10.5 3.2 13.8 1.1 27.5
Marital status
Never married 13.1 14.2 8.9 1.6 14.9 0.5 32.8 Married or living together 3.7 3.9 6.1 1.8 13.9 0.8 20.7 Divorced/separated/widowed 3.3 2.7 3.5 2.1 14.2 0.9 18.5
Residence
Urban 4.1 3.6 6.6 1.9 14.4 0.7 21.5 Rural 4.0 4.7 4.7 1.8 13.5 1.0 20.1
68
Kept from
seeing
friends or
family
Stopped from
leaving home
Asked to
report
activities
Stalked Rumour,
photos or
videos spread
Forced to
have an
abortion
Any social
violence
Region
Western 4.7 6.6 3.9 1.1 15.1 2.2 23.3 Central 5.0 4.6 3.9 0.3 8.5 2.1 16.8 Greater Accra 3.1 3.2 6.0 2.5 14.0 0.1 19.5 Volta 5.4 6.2 6.8 2.3 19.2 0.4 24.9 Eastern 5.2 5.5 7.6 2.9 25.9 1.1 32.3 Ashanti 3.2 2.8 6.8 1.5 16.3 0.4 23.5 Brong Ahafo 1.8 1.4 6.1 2.6 7.2 1.0 13.3 Northern 7.2 4.9 2.9 2.0 4.7 0.2 13.9 Upper East 1.8 6.7 0.6 0.6 9.6 0.0 14.5 Upper West 8.0 5.6 10.2 1.1 8.9 1.1 18.1
Education level
None 2.4 2.2 1.7 1.3 8.5 0.3 12.5 Primary 4.1 4.9 4.9 1.5 17.9 0.7 23.6 Middle/JSS/JHS 5.0 5.3 5.7 2.0 16.3 1.3 23.7 Secondary 4.5 4.6 13.1 3.4 13.9 0.5 26.9 Technical 0.0 1.3 7.5 1.7 11.3 0.6 15.8 Higher 4.8 2.4 7.4 1.2 10.4 0.8 17.0
Asset quintile
Lowest 4.4 4.5 3.9 1.6 14.0 0.9 20.2 Second 4.3 4.5 5.1 2.4 13.3 1.0 20.2 Middle 3.7 4.6 5.8 1.9 14.7 1.6 22.6 Fourth 4.0 4.1 7.7 1.9 14.7 0.4 21.2 Highest 3.6 2.5 6.0 1.2 13.3 0.2 20.4
Total 4.1 4.1 5.7 1.9 14.0 0.8 20.8
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
69
Table 8: Percentage of men who have experienced domestic or non-domestic social violence over their lifetime
Kept from
seeing
friends or
family
Stopped from
leaving home
Asked to
report
activities
Stalked Rumour,
photos or
videos spread
Forced to
have an
abortion
Any social
violence
Age group (years)
15–19 11.8 19.0 5.5 2.5 18.5 N/A 39.8 20–24 11.4 10.0 15.0 4.5 24.1 N/A 36.6 25–29 8.6 10.1 14.0 4.4 22.4 N/A 36.3 30–39 6.8 9.2 10.5 2.5 24.6 N/A 33.9 40–49 7.9 8.0 4.3 2.9 22.5 N/A 29.7 50–60 9.6 11.1 4.3 2.0 22.4 N/A 32.2
Employment
Self-employed 9.9 11.6 7.8 2.1 22.1 N/A 32.7 Employed 7.1 8.2 13.1 5.2 26.2 N/A 37.6 Not working 8.8 11.2 8.5 3.1 20.4 N/A 34.6
Marital status
Never married 10.6 13.6 5.4 2.1 15.5 N/A 30.0 Married or living together 9.0 10.5 9.7 2.8 24.0 N/A 35.1 Divorced/separated/widowed 7.0 8.1 12.2 5.6 25.1 N/A 35.8
Residence
Urban 7.7 8.4 10.0 3.6 22.5 N/A 33.8 Rural 10.1 12.6 8.8 2.8 23.1 N/A 35.0
70
Kept from
seeing
friends or
family
Stopped from
leaving home
Asked to
report
activities
Stalked Rumour,
photos or
videos spread
Forced to
have an
abortion
Any social
violence
Region
Western 5.9 9.3 10.7 6.3 34.8 N/A 49.3 Central 10.1 9.6 9.0 4.1 25.1 N/A 35.0 Greater Accra 5.3 3.6 9.6 2.6 12.0 N/A 22.9 Volta 2.2 5.7 7.6 1.7 33.5 N/A 39.6 Eastern 28.4 33.1 10.4 4.4 34.7 N/A 59.2 Ashanti 4.1 7.9 11.8 3.9 24.3 N/A 33.9 Brong Ahafo 11.1 11.0 12.8 3.1 21.4 N/A 33.0 Northern 13.2 14.3 4.9 0.6 11.9 N/A 21.5 Upper East 12.2 11.8 4.0 1.0 6.1 N/A 17.4 Upper West 6.1 6.1 1.8 0.0 14.9 N/A 19.2
Education level
None 10.6 14.5 3.0 0.5 14.1 N/A 25.6 Primary 12.7 14.4 5.8 1.3 25.4 N/A 37.0 Middle/JSS/JHS 8.1 11.1 7.7 2.5 23.2 N/A 34.6 Secondary 8.4 9.3 13.4 6.6 29.6 N/A 41.0 Technical 7.8 1.2 9.0 4.0 17.9 N/A 26.2 Higher 7.1 6.3 17.4 4.1 19.6 N/A 32.8
Asset quintile
Lowest 12.0 13.9 5.7 1.9 20.2 N/A 31.5 Second 9.8 11.9 7.5 2.5 22.3 N/A 34.5 Middle 7.3 8.9 8.6 2.2 26.9 N/A 37.1 Fourth 7.6 10.0 12.0 4.0 24.3 N/A 35.6 Highest 7.0 6.9 14.6 5.7 21.0 N/A 33.7
Total 8.9 10.6 9.4 3.2 22.8 N/A 34.4
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
71
Table 9: Percentage of men who experienced domestic or non-domestic social violence in the last 12 months
Kept from
seeing
friends or
family
Stopped from
leaving home
Asked to
report
activities
Stalked Rumour,
photos or
videos spread
Forced to
have an
abortion
Any social
violence
Age group (years)
15–19 5.3 7.1 4.1 1.7 13.0 N/A 23.1 20–24 2.3 1.0 10.6 3.5 16.4 N/A 26.1 25–29 1.0 1.4 8.1 3.5 12.1 N/A 20.5 30–39 0.6 1.5 5.7 1.5 13.3 N/A 18.2 40–49 0.8 0.7 2.1 1.8 11.2 N/A 13.9 50–60 0.8 1.8 2.2 2.0 8.5 N/A 12.1
Employment
Self-employed 1.2 1.4 4.3 1.8 11.9 N/A 16.2 Employed 1.2 1.6 7.7 3.3 15.0 N/A 22.7 Not working 2.5 3.4 6.2 2.4 10.8 N/A 19.9
Marital status
Never married 2.5 4.0 3.1 1.5 8.8 N/A 15.1 Married or living together 1.4 1.7 5.8 1.8 12.9 N/A 18.7 Divorced/separated/widowed 0.8 0.5 7.6 4.9 14.5 N/A 22.3
Residence
Urban 1.7 2.3 6.4 2.8 13.6 N/A 21.5 Rural 1.3 1.5 4.9 1.9 11.4 N/A 16.1
72
Kept from
seeing
friends or
family
Stopped from leaving home
Asked to
report
activities
Stalked Rumour,
photos or videos spread
Forced to
have an
abortion
Any social
violence
Region
Western 1.8 4.3 7.7 5.6 17.8 N/A 30.4 Central 2.3 2.0 6.9 3.3 15.8 N/A 19.7 Greater Accra 1.9 1.8 7.1 2.6 6.9 N/A 15.8 Volta 0.5 1.3 5.6 1.1 18.3 N/A 24.1 Eastern 1.8 2.1 5.1 2.5 20.6 N/A 24.5 Ashanti 1.1 1.0 8.2 2.9 13.4 N/A 20.6 Brong Ahafo 1.4 1.8 2.0 0.9 12.9 N/A 16.0 Northern 0.8 1.1 1.4 0.3 2.3 N/A 4.6 Upper East 0.7 0.7 1.0 0.0 3.1 N/A 4.8 Upper West 4.4 4.4 0.0 0.0 10.2 N/A 14.6
Education level
None 1.0 3.3 0.4 0.0 6.5 N/A 10.2 Primary 2.8 2.2 2.6 1.3 14.4 N/A 17.9 Middle/JSS/JHS 2.1 2.5 5.0 1.8 11.4 N/A 17.7 Secondary 0.5 0.9 7.6 4.6 18.1 N/A 24.4 Technical 3.1 0.0 9.0 4.0 10.5 N/A 21.6 Higher 0.3 0.5 11.3 3.3 11.3 N/A 21.6
Asset quintile
Lowest 1.2 2.6 2.5 1.7 9.8 N/A 13.3 Second 1.6 1.2 4.0 1.9 12.2 N/A 17.3 Middle 1.5 1.3 5.2 1.2 13.5 N/A 18.3 Fourth 1.9 2.1 7.5 2.0 13.9 N/A 22.5 Highest 1.1 2.5 10.1 5.3 13.5 N/A 23.6
Total 1.5 1.9 5.6 2.3 12.5 N/A 18.7
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
73
4.1.2. Physical violence
Men reported higher incidence levels of physical violence than women (Table 5): 42.4 per
cent of female respondents and 50.8 per cent of male respondents experienced at least one
of the above acts of physical violence over their lifetime. Almost 9 per cent of women and
10.2 per cent of men experienced at least one act of physical violence during the 12 months
prior to the survey. Tables 10–13 show further disaggregated detail on the incidence of
physical violence across different population groups and characteristics, which are
discussed in the paragraphs below.
Dimensions of physical violence. The most common forms of physical violence
experienced by both men and women over their lifetime were being slapped or being hit
by thrown objects. This form of violence affected 32.7 per cent of women (Table 10) and
38.3 per cent of men (Table 12). The next most prominent form of physical violence was
being hit by another person, experienced by 15.1 per cent of women (Table 10) and 20.8 per
cent of men (Table 12). These patterns of physical violence were very similar across the
lifetime of individuals and in the 12 months prior to the survey.
Age patterns. The incidence of physical violence over people’s lifetimes was strongly
associated with age for both women and men, with older individuals less likely to have
experienced physical violence. Over 47 per cent of women aged 24 years and below
reported having experienced physical violence, compared to 40.6 per cent of women aged
40–49 years and 35.7 per cent of women aged 50–60 years (Table 10). Younger men aged
15–19 years were particularly at risk of experiencing physical violence: 64.3 per cent of men
in this age group reported having experienced at least one form of physical violence over
their lifetime, compared to, for instance, 42.9 per cent of men aged 50–60 years (Table 12).
These age patterns were very similar across the lifetime of individuals and in the 12 months
prior to the survey (Tables 11 and 13).
Employment status. Employed and non-working women were more likely to have
experienced physical violence over their lifetime (43.7 and 43.6 per cent, respectively), than
self-employed women (41.8 per cent) (Table 10). This pattern was similar among the male
sample, but incidence levels were higher: 56.9 per cent of men who were not working
reported having experienced at least one form of physical violence over their lifetime,
while 55 per cent of employed and 46 per cent of self-employed men reported experiencing
this (Table 12). The employment patterns were similar to those observed when considering
experiences of physical violence in the 12 months prior to the survey (Table 11 and 13).
Marital status. The lifetime incidence of physical violence was higher among individuals,
particularly women, who were divorced, separated or widowed, or married. Experiences
with at least one form of physical violence were reported by 46.3 per cent of divorced,
separated or widowed women and by 41.8 per cent of women who were married (or living
with a partner), in contrast to 35.1 per cent of women who were never married (Table 10).
The differences in the results for the male sample were less pronounced (Table 12): 52.6 per
74
cent of divorced, separated or widowed men reported having experienced physical
violence, compared to 51.4 per cent of men who were never married and 50.1 per cent who
were married. The incidence of physical violence in the 12 months prior to the survey was
highest among never-married women (Table 11).
Geographical patterns. The incidence of lifetime physical violence was higher among
women living in rural areas (44.4 per cent) than women living in urban areas (40.7 per cent)
(Table 10). A similar proportion of men in rural and urban areas experienced physical
violence (50.7 per cent and 50.9 per cent, respectively) (Table 12). Women and men in the
Eastern region reported having experienced higher levels of physical violence than those
in any other region. These patterns were common across the lifetime of individuals and in
the 12 months prior to the survey (Table 11 and 13).
Education levels and asset ownership. Women and men with no formal education (34.6
and 31.7 per cent, respectively) and women and men with higher education (38.0 and 51.2
per cent, respectively) were less likely to experience physical violence than women and
men with primary education (45.3 and 52.1 per cent, respectively) or other levels of
education (Tables 10 and 12). This suggests that intermediate levels of education may be
associated with increased risk of physical violence among women and men. Women in the
highest and fourth asset quintiles were less likely to experience physical violence (38.9 and
41.4 per cent, respectively) than women in the three lowest asset quintiles. For instance,
44.7 per cent of women in the lowest asset quintile reported experiencing physical violence
(Table 10). The opposite was true for men: men in the lowest asset quintile were less likely
to experience physical violence (46.1 per cent) than men in the highest asset quintile (50.1
per cent) (Table 12). These education and wealth patterns were very similar across the
lifetime of individuals and in the 12 months prior to the survey (Tables 11 and 13).
75
Table 10: Percentage of women who have experienced domestic or non-domestic physical violence over their lifetime
Slapped or
thrown
things at
Pushed
or shoved
Hit Kicked,
dragged or
beaten up
Choked or
strangled
Burnt Attacked
with
weapon
Any physical
violence
Age group (years)
15–19 35.0 18.5 13.5 16.8 2.4 1.7 1.3 47.6 20–24 36.2 15.0 14.7 15.2 1.7 1.2 2.2 47.0 25–29 34.7 18.9 14.5 14.2 2.7 1.7 2.6 45.5 30–39 32.0 13.3 15.3 13.7 2.1 0.7 4.1 42.0 40–49 32.1 14.6 17.0 12.7 2.3 1.0 2.8 40.6 50–60 28.5 12.3 14.2 11.5 2.1 0.5 1.7 35.7
Employment
Self-employed 32.8 14.5 16.1 13.6 2.6 1.1 2.8 41.8 Employed 29.6 17.5 14.4 13.6 1.8 0.3 3.9 43.7 Not working 33.9 15.1 12.6 14.0 1.4 1.0 2.3 43.6
Marital status
Never married 22.3 17.9 10.4 9.4 0.0 1.3 1.3 35.1 Married or living together 32.2 14.1 15.2 13.7 2.1 0.9 2.8 41.8 Divorced/separated/widowed 36.9 17.2 15.5 14.5 3.3 1.2 3.3 46.3
Residence
Urban 31.6 14.0 15.4 11.6 1.8 0.8 2.9 40.7 Rural 34.0 16.0 14.7 16.0 2.7 1.2 2.7 44.4
76
Slapped or
thrown
things at
Pushed
or shoved
Hit Kicked,
dragged or
beaten up
Choked or
strangled
Burnt Attacked
with
weapon
Any physical
violence
Region
Western 35.5 8.8 5.6 9.0 0.4 0.0 2.3 40.5 Central 36.0 11.4 16.0 13.0 5.2 0.5 3.7 44.2 Greater Accra 21.1 12.1 13.1 5.1 1.1 0.9 2.9 29.8 Volta 44.4 31.0 29.0 22.7 2.5 1.0 0.8 56.3 Eastern 42.8 25.3 29.4 28.2 4.9 3.9 4.6 61.9 Ashanti 37.6 11.9 11.0 16.5 1.3 0.4 3.3 46.3 Brong Ahafo 35.5 16.1 15.9 10.2 3.0 1.5 3.1 44.2 Northern 25.9 15.8 12.4 17.0 0.5 0.4 1.3 39.0 Upper East 11.2 5.5 3.2 4.4 2.4 1.1 0.0 13.5 Upper West 13.9 4.3 6.1 7.4 3.1 0.0 2.4 19.9
Education level
None 27.0 14.5 14.6 12.7 2.5 0.5 1.8 34.6
Primary 35.5 15.9 14.6 17.0 3.3 1.8 4.0 45.3 Middle/JSS/JHS 37.3 15.5 16.9 13.8 2.3 1.2 2.9 47.0 Secondary 28.2 12.5 13.2 13.0 1.1 1.2 1.6 40.3 Technical 34.8 19.5 17.5 10.3 0.0 0.0 5.9 45.2 Higher 25.5 13.5 9.9 10.8 0.9 0.0 3.6 38.0
Asset quintile
Lowest 36.1 20.2 18.7 15.6 3.9 0.9 2.3 44.7 Second 32.7 14.6 11.9 15.4 2.7 1.1 3.1 44.0 Middle 33.6 13.8 14.1 13.1 1.5 0.9 2.3 42.9 Fourth 29.0 12.5 17.4 12.5 1.3 1.0 2.5 38.9 Highest 33.0 13.9 13.7 10.6 1.6 1.2 4.0 41.4
Total 32.7 15.0 15.1 13.7 2.2 1.0 2.8 42.4
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
77
Table 11: Percentage of women who experienced domestic or non-domestic physical violence in the last 12 months
Slapped or
thrown
things at
Pushed
or shoved
Hit Kicked,
dragged or
beaten up
Choked or
strangled
Burnt Attacked
with
weapon
Any
physical
violence
Age group (years)
15–19 17.7 7.8 6.1 7.8 1.7 1.1 1.1 28.0 20–24 10.1 3.1 4.1 2.5 0.2 0.3 0.0 14.1 25–29 6.5 3.9 2.2 2.6 1.4 0.3 1.0 9.5 30–39 6.2 1.2 1.7 2.5 0.6 0.0 1.6 9.2 40–49 3.2 0.8 1.1 0.5 0.2 0.0 0.2 3.8 50–60 1.0 0.0 0.2 0.0 0.0 0.0 0.0 1.2
Employment
Self-employed 5.0 1.5 1.7 1.8 0.6 0.0 0.7 7.2 Employed 4.8 2.4 2.2 1.4 0.4 0.0 0.7 7.0 Not working 10.0 3.7 3.1 3.3 0.7 0.7 1.1 14.5
Marital status
Never married 11.7 6.4 3.5 4.5 0.0 1.3 1.3 20.0 Married or living together 5.9 1.6 2.0 1.6 0.6 0.1 0.7 8.2 Divorced/separated/widowed 5.6 2.9 2.0 3.2 0.9 0.0 0.8 8.5
Residence
Urban 6.0 1.7 2.3 2.1 0.5 0.1 0.9 8.6 Rural 6.4 2.6 1.7 2.2 0.7 0.2 0.6 9.3
78
Slapped or
thrown
things at
Pushed or
shoved
Hit Kicked,
dragged or
beaten up
Choked or
strangled
Burnt Attacked
with
weapon
Any
physical
violence
Region
Western 6.1 1.6 1.3 1.3 0.0 0.0 0.6 8.0 Central 6.3 1.4 2.9 2.8 0.8 0.5 1.6 10.9 Greater Accra 4.9 1.9 1.8 1.5 0.2 0.0 1.2 6.5 Volta 8.1 2.7 2.8 2.7 0.3 1.0 0.0 10.3 Eastern 8.1 2.3 3.1 3.1 1.4 0.0 1.2 11.1 Ashanti 7.3 2.0 1.8 2.2 0.3 0.0 0.4 10.4 Brong Ahafo 6.1 3.3 2.7 2.2 1.2 0.2 0.7 9.5 Northern 4.5 2.3 0.6 2.0 0.5 0.0 0.0 7.4 Upper East 1.8 1.8 0.8 0.8 1.8 0.0 0.0 1.8 Upper West 2.4 1.6 1.7 2.3 1.6 0.0 1.2 7.6
Education level
None 3.5 1.3 1.1 0.6 0.6 0.0 0.3 4.3
Primary 7.7 3.6 2.6 4.0 1.2 0.2 0.8 12.1 Middle/JSS/JHS 7.0 2.0 2.3 2.4 0.5 0.3 0.9 10.2 Secondary 6.9 2.6 3.2 2.2 0.5 0.0 0.5 10.9 Technical 7.6 3.5 1.9 4.0 0.0 0.0 3.5 9.5 Higher 3.4 0.2 0.2 0.3 0.0 0.0 0.5 4.1
Asset quintile
Lowest 6.5 3.1 3.1 2.4 1.6 0.4 0.3 10.0 Second 6.5 2.2 1.7 2.3 0.5 0.1 0.8 8.9 Middle 7.8 1.7 2.1 2.8 0.2 0.0 1.0 12.1 Fourth 4.7 1.6 2.0 1.1 0.5 0.2 0.3 6.4 Highest 5.5 1.7 1.4 2.2 0.0 0.0 1.6 7.8
Total 6.2 2.1 2.1 2.1 0.6 0.2 0.7 8.9
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
79
Table 12: Percentage of men who have experienced domestic or non-domestic physical violence over their lifetime
Slapped or
thrown
things at
Pushed or
shoved
Hit Kicked,
dragged or
beaten up
Choked or
strangled
Burnt Attacked
with
weapon
Any
physical
violence
Age group (years)
15–19 47.2 30.9 22.8 15.9 3.7 2.1 3.7 64.3 20–24 38.6 17.9 20.1 16.9 1.4 3.1 5.1 54.2 25–29 37.4 16.5 16.9 12.0 3.6 2.1 6.9 47.5 30–39 37.9 18.3 21.6 14.5 4.6 1.6 10.2 50.9 40–49 39.0 20.8 22.6 16.7 3.3 1.3 8.3 49.5 50–60 32.6 22.6 21.9 17.5 3.3 1.9 3.8 42.9
Employment
Self-employed 34.3 18.5 18.5 14.3 3.4 1.6 6.8 46.0 Employed 42.1 21.9 22.9 16.8 3.8 2.6 9.6 55.1 Not working 43.2 22.4 23.8 16.1 2.8 2.1 3.7 56.9
Marital status
Never married 36.6 20.9 17.7 15.4 2.0 1.9 3.0 51.4 Married or living together 37.4 20.4 21.6 16.2 3.9 1.7 7.9 50.1 Divorced/separated/widowed 42.8 18.5 20.7 12.3 3.0 3.1 7.1 52.6
Residence
Urban 38.3 19.0 19.8 13.7 3.9 2.0 8.0 50.9 Rural 38.2 21.4 21.8 16.9 2.9 1.9 5.9 50.7
80
Slapped or
thrown
things at
Pushed or
shoved
Hit Kicked,
dragged or
beaten up
Choked or
strangled
Burnt Attacked
with
weapon
Any
physical
violence
Region
Western 46.1 16.7 17.3 13.9 3.8 1.4 12.2 60.0 Central 41.4 19.9 21.7 6.7 4.5 3.6 7.0 52.6 Greater Accra 26.7 15.2 17.9 5.8 2.8 2.1 5.0 35.5 Volta 44.6 33.4 35.5 32.1 1.8 0.0 2.5 59.5 Eastern 44.2 28.6 27.5 20.3 3.9 3.2 7.3 66.1 Ashanti 49.8 20.4 23.1 23.1 3.9 2.9 9.4 63.0 Brong Ahafo 35.0 20.4 19.1 6.7 6.2 3.6 10.2 48.7 Northern 31.3 16.2 12.9 18.7 0.0 0.0 3.3 41.6 Upper East 16.3 13.0 10.1 6.0 5.2 0.0 0.4 19.1 Upper West 14.3 14.3 7.1 1.5 3.2 0.0 7.0 21.7
Education level
None 23.8 15.4 15.6 12.3 2.8 1.4 3.2 31.7 Primary 39.7 23.3 20.7 13.2 2.2 1.6 6.3 52.1 Middle/JSS/JHS 39.4 20.0 20.8 15.6 4.1 2.3 7.5 52.1 Secondary 44.4 21.3 25.0 16.0 3.7 2.3 6.7 59.2 Technical 44.4 19.7 19.9 14.8 0.0 1.4 6.8 57.8 Higher 38.3 21.2 20.1 18.8 3.7 1.5 9.8 51.2
Asset quintile
Lowest 37.2 22.0 20.8 15.4 3.2 1.2 3.3 46.1 Second 36.3 20.3 22.5 16.7 2.9 2.7 5.9 51.4 Middle 40.9 18.3 22.6 15.4 4.6 1.6 7.8 53.1 Fourth 41.4 19.1 19.0 15.9 4.1 3.1 8.0 53.4 Highest 36.5 21.1 18.6 12.3 2.3 1.0 10.7 50.1
Total 38.3 20.2 20.8 15.3 3.4 2.0 6.9 50.8
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
81
Table 13: Percentage of men who experienced domestic or non-domestic physical violence in the last 12 months
Slapped or
thrown
things at
Pushed or
shoved
Hit Kicked,
dragged or
beaten up
Choked or
strangled
Burnt Attacked
with
weapon
Any
physical
violence
Age group (years)
15–19 18.6 14.9 7.8 3.3 3.1 0.8 2.0 30.6 20–24 7.2 3.0 3.0 2.1 0.8 0.5 1.0 14.2 25–29 4.9 2.0 1.1 0.4 1.5 0.0 0.4 8.7 30–39 3.9 2.4 1.6 0.8 0.3 0.0 1.7 7.3 40–49 2.1 1.8 2.2 0.5 0.5 0.0 0.5 5.1 50–60 1.0 1.4 1.2 0.3 0.3 0.8 0.7 4.0
Employment
Self-employed 3.0 1.7 1.9 0.7 0.7 0.2 1.0 6.5 Employed 5.2 3.2 1.7 0.7 0.6 0.0 1.1 9.9 Not working 11.4 8.1 4.5 2.3 1.9 0.7 1.1 19.4
Marital status
Never married 10.9 8.4 5.1 1.8 1.1 0.9 0.3 18.6 Married or living together 3.6 2.3 1.8 1.0 0.8 0.2 1.1 7.7 Divorced/separated/widowed 6.6 2.8 2.2 0.5 1.1 0.0 1.6 11.0
Residence
Urban 5.0 3.8 2.2 0.6 0.7 0.6 0.8 9.7 Rural 5.7 3.1 2.6 1.5 1.1 0.0 1.2 10.6
82
Slapped
or thrown
things at
Pushed or
shoved
Hit Kicked,
dragged or
beaten up
Choked or
strangled
Burnt Attacked
with
weapon
Any
physical
violence
Region
Western 9.0 1.8 2.2 0.7 0.6 0.0 1.6 12.5 Central 9.0 3.4 1.8 1.8 1.4 1.2 1.4 14.9 Greater Accra 4.3 4.3 1.7 0.0 1.0 0.0 0.9 8.4 Volta 3.4 1.5 3.0 2.7 0.0 0.0 0.4 9.1 Eastern 6.1 4.8 3.6 0.0 0.0 0.0 0.6 12.3 Ashanti 6.2 4.4 3.3 2.6 1.9 1.0 1.5 14.0 Brong Ahafo 5.0 4.3 2.2 0.0 1.8 0.0 1.5 8.4 Northern 2.0 1.4 0.7 0.0 0.0 0.0 0.4 3.1 Upper East 3.5 4.9 3.1 1.1 1.1 0.0 0.0 5.6 Upper West 2.2 3.8 1.7 0.0 0.0 0.0 1.7 6.1
Education level
None 3.1 1.2 1.2 0.3 0.6 0.0 0.0 3.8 Primary 5.3 7.2 3.6 1.8 1.2 0.7 1.0 11.1 Middle/JSS/JHS 5.6 3.3 2.8 1.1 0.9 0.3 1.7 11.0 Secondary 8.7 4.4 3.4 1.9 1.6 0.4 1.3 16.2 Technical 6.3 3.4 0.0 0.0 0.0 0.0 0.6 9.1 Higher 2.3 1.6 0.7 0.0 0.3 0.0 0.3 5.0
Asset quintile
Lowest 7.3 2.9 2.0 0.6 1.1 0.0 1.3 10.8 Second 6.9 4.3 2.5 1.4 1.6 0.3 1.1 13.1 Middle 4.1 2.5 2.0 1.2 0.4 0.0 0.7 6.8 Fourth 3.3 2.5 2.8 1.8 0.5 1.0 0.5 9.1 Highest 4.4 4.9 2.6 0.0 0.6 0.0 1.6 9.6
Total 5.4 3.5 2.4 1.0 0.9 0.3 1.0 10.2
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
83
4.1.3. Sexual violence
Table 5 shows that 30 per cent of women and 23.1 per cent of men experienced sexual
violence at least once over their lifetime, and 10.6 per cent of women and 9.5 per cent of
men reported having experienced sexual violence at least once over the 12 months that
preceded the survey. Tables 14 to 17 provide further disaggregation of the data on the
incidence of sexual violence, as discussed below.
Dimensions of sexual violence. The most prevalent form of sexual violence among both
women and men over their lifetimes was being subject to sexual comments (18 per cent
among women and 12.1 per cent among men) (Tables 14 and 16). This was followed by
unwanted touches (15.3 per cent among women and 10.3 per cent among men) and being
physically forced to have sex (9 per cent among women and 6.9 per cent among men).
These patterns of sexual violence were very similar across the lifetime of individuals and
in the 12 months prior to the survey (Tables 15 and 17).
Age patterns. Similar to social and physical violence, the incidence of lifetime sexual
violence was higher among younger women: 38.2 per cent of women aged 15–19 years,
40.4 per cent of women aged 20–24 years and 38.3 per cent of women aged 30–39 years
reported having experienced at least one act of sexual violence. In contrast, 18.8 per cent of
women aged 50–60 years experienced these forms of violence (Table 14). The incidence of
sexual violence was highest among men aged 30–39 years (29.7 per cent) and men aged 20–
24 years (29.2 per cent). The lowest incidence was again among the oldest male age group
(13.8 per cent) (Table 16). These age patterns were also observed when considering the 12
months prior to the survey (Tables 15 and 17).
Employment status. Employed women and men (41.9 and 31.5 per cent, respectively) were
more likely to have experienced sexual violence than men and women not in employment
(Tables 14 and 16). This is an issue that will be explored further when focusing the analysis
on domestic violence. These patterns were very similar across the lifetime of individuals
and in the 12 months prior to the survey (Tables 15 and 17).
Marital status. As observed with social and physical violence, the incidence of sexual
violence was more prominent among women and men who were divorced, separated or
widowed at the time of the survey (31.7 per cent of women and 29.4 per cent of men) (Table
14). The differences in relation to other marital status were considerable, with 22.6 per cent
of never-married women and 16.9 per cent of never-married men having reported sexual
violence (Table 16). As with other forms of violence, the incidence of sexual violence in the
12 months prior to the survey was highest among never-married women (Table 17). This
may indicate a generational effect, which will be explored further in the next sections.
Geographical patterns. Contrary to other forms of violence analysed above, the incidence
of sexual violence was consistently higher among women and men living in urban areas
(32.1 per cent of women and 28.5 per cent of men, compared with 27.8 and 18.1 per cent in
84
rural areas, respectively) (Tables 14 and 16). As with other forms of violence, the incidence
of sexual violence among women was higher in the Eastern region (45.5 per cent) than in
any other region. The highest levels of incidence of sexual violence among men were
observed in the Ashanti region (32.4 per cent). The lowest incidence of sexual violence
against women and men was in the Upper East region (8.8 per cent and 6.7 per cent,
respectively). These geographical patterns were very similar when considering the
incidence of sexual violence in the 12 months prior to the survey (Tables 15 and 17).
Education levels and asset ownership. Women with no formal education were about half
as likely as women with some level of education to report sexual violence. Women with
secondary (39.1 per cent) and higher education (37.1 per cent) were more likely to have
experienced sexual violence than women with primary (32.8 per cent) or middle levels of
education (32.0 per cent) (Table 14). Men with higher levels of education were more likely
to have experienced sexual violence over their lifetime: 34.9 per cent, compared to 20.1 per
cent among men with primary education. Men with no formal education were the least
likely to report sexual violence over their lifetime (7.8 per cent) (Table 16). Reflecting this
latter pattern, sexual violence against men was also more prominent at the top end of the
asset ownership distribution. The proportion of men experiencing any type of sexual
violence over their lifetime was twice as high for male respondents in the highest asset
quintile (29.3 per cent) than for those in the lowest quintile (15.1 per cent) (Table 16). This
association was similar for women, but the differences across quintiles were less
pronounced (Table 14). These patterns were also observed when considering the incidence
of sexual violence in the 12 months prior to the survey (Tables 15 and 17).
85
Table 14: Percentage of women who have experienced domestic or non-domestic sexual violence over their lifetime
Sexual
comment
Sexual
touch
Physically
forced to
have sex
Otherwise
forced to
have sex
Sex
without
consent
Sex
without
protection
Sex
because
afraid
Any
sexual
violence
Age group (years)
15–19 20.6 26.4 10.3 3.5 3.3 2.0 0.9 38.2 20–24 25.6 18.6 11.8 1.4 7.9 3.4 5.1 40.4 25–29 19.8 20.2 14.0 2.1 6.6 4.1 6.1 38.3 30–39 17.8 14.2 8.9 1.6 5.9 2.7 4.6 29.5 40–49 13.8 10.5 6.2 1.4 3.5 0.5 4.1 21.4 50–60 13.8 9.6 3.2 1.6 2.1 1.3 2.5 18.8
Employment
Self-employed 16.4 13.8 7.9 1.6 4.8 2.2 4.2 27.0 Employed 26.2 21.2 15.2 1.9 6.7 4.4 4.6 41.9 Not working 18.8 16.8 9.2 2.1 5.4 1.9 4.5 32.9
Marital status
Never married 14.6 14.7 2.0 0.9 0.9 0.0 0.0 22.6 Married or living together 18.0 14.9 9.2 1.7 5.2 2.3 4.6 30.1 Divorced/separated/widowed 18.9 16.9 10.0 2.3 5.8 3.4 4.3 31.7
Residence
Urban 19.0 16.0 10.1 1.6 5.3 2.9 5.2 32.1 Rural 16.9 14.5 7.8 1.9 4.9 1.8 3.2 27.8
86
Sexual
comment
Sexual
touch
Physically
forced to
have sex
Otherwise
forced to
have sex
Sex
without
consent
Sex
without
protection
Sex
because
afraid
Any
sexual
violence
Region
Western 15.2 13.0 9.3 2.3 2.9 1.8 1.6 26.7 Central 19.7 19.6 8.8 1.4 8.3 6.2 6.6 35.5 Greater Accra 14.7 12.9 9.3 1.0 6.9 3.0 7.1 28.1 Volta 17.5 13.5 9.0 4.4 5.7 0.6 8.5 29.9 Eastern 26.7 26.8 14.6 2.2 7.0 5.0 6.9 45.5 Ashanti 19.9 14.5 8.6 2.1 4.3 0.9 1.7 30.5 Brong Ahafo 24.7 16.6 9.9 0.9 4.3 1.8 1.9 32.7 Northern 10.9 13.4 3.0 0.6 2.0 0.3 0.6 19.5 Upper East 4.7 4.4 5.0 0.0 1.1 1.7 0.7 8.8 Upper West 6.0 3.6 3.5 2.1 1.1 1.1 3.5 12.9
Education level
None 10.6 8.7 3.9 1.3 3.4 1.1 2.6 17.4 Primary 19.9 18.2 10.5 2.0 6.4 2.6 4.6 32.8 Middle/JSS/JHS 20.0 16.6 9.8 1.9 5.3 2.3 4.4 32.0 Secondary 18.9 18.1 11.6 1.9 7.2 3.5 5.8 39.1 Technical 20.1 15.8 11.7 1.2 4.6 3.6 8.3 35.2 Higher 25.7 18.8 12.2 1.5 3.3 4.2 3.9 37.1
Asset quintile
Lowest 15.6 15.7 7.5 0.9 4.3 1.8 3.9 27.6 Second 17.0 15.4 8.6 2.1 4.1 2.1 3.8 27.8 Middle 20.1 16.3 10.4 2.6 6.1 2.6 4.3 32.1 Fourth 17.4 14.0 8.5 1.7 5.0 3.1 4.6 31.2 Highest 21.5 15.6 10.7 1.3 7.2 2.5 5.1 33.3
Total 18.0 15.3 9.0 1.7 5.1 2.4 4.3 30.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
87
Table 15: Percentage of women who experienced domestic or non-domestic sexual violence in the last 12 months
Sexual
comment
Sexual
touch
Physically
forced to
have sex
Otherwise
forced to
have sex
Sex
without
consent
Sex
without
protection
Sex
because
afraid
Any
sexual
violence
Age group (years)
15–19 11.9 15.4 3.1 0.9 0.4 0.5 1.6 22.1 20–24 13.1 8.1 2.2 0.1 1.6 1.8 1.1 20.2 25–29 7.3 6.1 1.8 0.5 0.9 1.8 1.4 13.7 30–39 7.0 2.5 0.8 0.2 0.7 1.1 0.8 10.1 40–49 2.3 2.0 0.2 0.0 0.1 0.4 0.0 4.2 50–60 1.9 0.9 0.0 0.3 0.0 0.0 0.0 2.1
Employment
Self-employed 4.8 3.2 1.1 0.2 0.7 0.7 0.6 8.2 Employed 11.9 5.7 0.9 0.0 0.0 1.7 1.1 16.0 Not working 8.9 7.0 1.0 0.5 0.6 1.4 0.8 14.7
Marital status
Never married 7.6 8.5 0.3 0.3 0.0 0.0 0.0 12.8 Married or living together 5.8 3.9 1.3 0.3 0.8 1.2 0.8 10.0 Divorced/separated/widowed 8.9 5.2 0.5 0.2 0.1 0.3 0.7 12.2
Residence
Urban 7.6 4.3 1.1 0.1 0.5 1.5 1.0 11.7 Rural 5.4 4.5 1.1 0.4 0.7 0.5 0.4 9.3
88
Sexual
comment
Sexual
touch
Physically
forced to
have sex
Otherwise
forced to
have sex
Sex
without
consent
Sex
without
protection
Sex
because
afraid
Any
sexual
violence
Region
Western 5.8 4.2 1.5 0.0 0.3 0.4 0.6 7.9 Central 6.0 5.2 1.6 0.5 1.2 2.0 1.4 13.5 Greater Accra 7.1 4.4 0.9 0.1 1.0 2.2 1.3 11.5 Volta 2.5 2.6 1.7 0.6 0.0 1.0 0.3 7.1 Eastern 6.9 8.0 1.2 0.3 0.6 0.9 0.6 12.5 Ashanti 8.1 3.5 1.3 0.6 0.8 0.7 0.1 12.5 Brong Ahafo 11.7 6.2 0.3 0.0 0.3 0.0 1.0 14.7 Northern 2.3 2.2 0.8 0.0 0.3 0.0 0.0 3.1 Upper East 0.6 0.0 0.0 0.0 0.0 0.0 0.6 1.2 Upper West 3.6 2.4 0.0 0.0 0.0 0.0 1.1 5.9
Education level
None 1.8 1.6 0.5 0.4 0.1 0.0 0.1 3.6
Primary 6.6 5.2 1.1 0.0 0.7 1.2 0.1 10.9 Middle/JSS/JHS 7.8 5.1 1.3 0.2 0.9 0.9 0.9 12.2 Secondary 7.8 4.7 1.9 0.6 0.8 2.4 1.5 14.7 Technical 9.0 5.0 1.4 0.0 1.4 3.3 1.4 12.3 Higher 12.6 7.0 0.0 0.0 0.0 0.7 1.2 15.9
Asset quintile
Lowest 5.2 4.3 1.3 0.2 0.1 0.5 0.3 9.6 Second 5.9 5.0 1.0 0.1 0.7 0.9 0.6 9.9 Middle 7.9 5.2 1.7 0.4 0.4 0.4 0.7 11.2 Fourth 7.7 4.1 0.9 0.5 1.2 1.9 1.0 12.9 Highest 6.3 3.0 0.7 0.0 0.5 1.2 1.0 9.1
Total 6.6 4.4 1.1 0.3 0.6 1.0 0.7 10.6
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
89
Table 16: Percentage of men who have experienced domestic or non-domestic sexual violence over their lifetime
Sexual
comment
Sexual
touch
Physically
forced to
have sex
Otherwise
forced to
have sex
Sex
without
consent
Sex
without
protection
Sex
because
afraid
Any
sexual
violence
Age group (years)
15–19 14.0 5.8 2.6 0.8 0.8 0.0 0.7 19.4 20–24 17.3 15.5 9.9 2.4 2.6 0.5 3.7 29.2 25–29 11.4 10.1 7.0 3.4 3.3 2.0 3.8 25.1 30–39 14.6 14.0 8.0 2.7 2.8 1.4 3.7 29.7 40–49 7.0 7.6 6.4 0.5 1.6 2.2 1.3 16.5 50–60 8.0 5.0 5.6 1.2 1.4 0.2 0.8 13.8
Employment
Self-employed 9.3 7.7 5.6 0.9 1.6 1.0 1.8 18.3 Employed 13.6 15.4 10.9 4.8 4.7 1.8 5.7 31.5 Not working 17.1 10.1 5.3 1.3 0.9 0.9 0.8 24.6
Marital status
Never married 12.5 7.1 3.8 1.8 2.7 0.0 0.3 16.9 Married or living together 10.9 9.9 7.4 2.0 2.2 1.2 2.9 22.9 Divorced/separated/widowed 15.6 13.9 8.2 2.2 2.4 2.6 3.7 29.4
Residence
Urban 16.1 12.8 8.2 3.0 2.9 1.1 2.4 28.5 Rural 8.3 7.8 5.7 1.0 1.7 1.3 2.7 18.1
90
Sexual
comment
Sexual
touch
Physically
forced to
have sex
Otherwise
forced to
have sex
Sex
without
consent
Sex
without
protection
Sex
because
afraid
Any
sexual
violence
Region
Western 9.4 15.0 10.6 2.6 1.1 1.1 2.8 26.0 Central 12.8 5.7 7.2 1.7 7.4 4.9 3.7 24.2 Greater Accra 14.1 10.3 7.8 2.8 3.2 1.3 3.6 26.3 Volta 10.0 5.7 4.1 1.7 1.4 0.0 0.9 15.5 Eastern 14.7 15.9 9.0 2.1 4.6 1.4 7.6 29.8 Ashanti 17.8 14.1 9.7 2.3 2.2 1.5 2.0 32.4 Brong Ahafo 13.3 8.4 2.8 2.7 0.6 0.0 0.0 17.6 Northern 4.2 4.9 3.2 0.4 0.0 0.8 1.0 12.6 Upper East 3.0 5.7 2.0 0.0 0.0 0.0 1.1 6.7 Upper West 3.5 3.9 3.5 0.0 0.0 0.0 1.7 7.3
Education level
None 1.8 3.5 2.6 0.3 0.9 0.0 0.7 7.8 Primary 10.0 6.6 6.0 1.8 2.8 1.6 0.7 20.1 Middle/JSS/JHS 12.8 8.5 5.5 0.8 0.9 1.0 2.3 22.3 Secondary 16.4 12.2 8.5 3.1 2.7 1.2 4.7 28.7 Technical 13.2 8.3 13.0 4.2 6.0 2.2 5.9 24.4 Higher 15.9 22.3 12.1 4.8 5.2 2.6 3.1 34.9
Asset quintile
Lowest 7.6 5.5 4.5 1.3 0.9 1.3 0.8 15.1 Second 10.5 7.9 5.2 1.1 2.9 1.1 1.7 18.8 Middle 13.6 9.2 6.0 0.5 1.7 1.0 3.3 22.5 Fourth 16.3 16.6 11.6 3.5 2.3 1.9 3.4 32.6 Highest 13.6 13.5 8.0 4.2 3.4 0.7 4.6 29.3
Total 12.1 10.3 6.9 2.0 2.3 1.2 2.6 23.1
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
91
Table 17: Percentage of men who experienced domestic or non-domestic sexual violence in the last 12 months
Sexual
comment
Sexual
touch
Physically
forced to
have sex
Otherwise
forced to
have sex
Sex
without
consent
Sex
without
protection
Sex
because
afraid
Any
sexual
violence
Age group (years)
15–19 10.8 4.8 1.9 0.8 0.8 0.7 0.0 14.5 20–24 7.9 6.3 1.8 1.6 0.3 1.6 0.5 14.2 25–29 5.8 4.7 1.7 1.8 0.8 1.1 0.4 12.2 30–39 6.0 3.3 0.7 0.9 0.3 1.1 0.7 10.2 40–49 2.6 1.4 2.1 0.0 0.0 0.0 0.2 5.8 50–60 0.0 0.0 0.7 0.0 0.0 0.0 0.0 0.7
Employment
Self-employed 3.2 2.0 0.9 0.5 0.1 0.3 0.1 5.9 Employed 6.9 5.6 2.9 1.5 0.7 2.3 0.5 14.8 Not working 8.3 4.0 1.0 1.1 0.4 0.1 0.7 11.9
Marital status
Never married 8.4 4.3 1.3 1.3 1.0 0.0 0.0 11.5 Married or living together 4.4 2.9 1.3 0.7 0.1 0.7 0.3 8.2 Divorced/separated/widowed 5.5 4.5 2.1 1.3 0.7 1.8 0.7 12.5
Residence
Urban 7.8 4.3 1.7 1.4 0.5 0.7 0.3 12.8 Rural 2.9 2.5 1.1 0.4 0.2 0.9 0.4 6.4
92
Sexual
comment
Sexual
touch
Physically
forced to
have sex
Otherwise
forced to
have sex
Sex
without
consent
Sex
without
protection
Sex
because
afraid
Any
sexual
violence
Region
Western 0.3 2.0 2.8 0.6 0.0 0.7 0.7 6.5 Central 5.8 2.0 0.7 0.7 1.2 1.9 1.4 10.9 Greater Accra 7.3 4.7 1.8 2.1 0.4 1.3 0.3 13.2 Volta 2.9 2.7 1.5 0.9 0.5 0.5 0.0 5.7 Eastern 4.3 3.9 0.7 0.7 0.7 2.5 0.0 9.4 Ashanti 9.4 4.7 2.1 0.6 0.4 0.4 0.2 14.2 Brong Ahafo 8.6 5.0 0.9 1.5 0.0 0.0 0.0 10.5 Northern 2.7 1.1 0.0 0.0 0.0 0.0 0.8 4.2 Upper East 0.4 1.4 0.4 0.0 0.0 0.0 0.0 1.8 Upper West 0.0 2.2 1.7 0.0 0.0 0.0 0.0 3.8
Education level
None 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Primary 6.0 3.0 1.5 1.0 1.3 0.2 0.0 10.6 Middle/JSS/JHS 4.7 2.5 0.8 0.7 0.0 1.0 0.5 8.9 Secondary 7.8 6.4 2.1 1.7 0.2 1.4 0.3 13.7 Technical 8.4 5.9 3.8 2.7 2.2 0.0 0.0 11.7 Higher 7.2 4.4 3.0 0.5 0.4 0.8 0.8 13.1
Asset quintile
Lowest 4.1 2.0 1.3 0.7 0.2 0.6 0.7 6.8 Second 3.7 2.6 0.6 0.7 0.6 1.0 0.2 6.9 Middle 5.0 2.3 0.6 0.0 0.0 0.7 0.2 8.3 Fourth 7.8 5.6 3.1 2.0 0.4 0.7 0.3 13.9 Highest 6.6 4.8 1.8 0.9 0.4 1.0 0.4 13.2
Total 5.3 3.4 1.4 0.9 0.4 0.8 0.3 9.5
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
93
4.1.4. Psychological violence
Psychological violence was the most common type of violence experienced by women and
men across Ghana. About half of the respondents (51.9 per cent of men and 48 per cent of
women) experienced an act of psychological violence at least once in their lifetime (Table
5), while 22.5 per cent of women and 28 per cent of men experienced at least one type of
psychological violence in the 12 months prior to the survey (Table 6). Further
disaggregated information on the incidence of psychological violence is provided in Tables
18–21 and discussed in the paragraphs below.
Dimensions of psychological violence. The most common type of psychological violence
reported by both men and women over their lifetime was being insulted, humiliated or
belittled. Almost 46 per cent of women (Table 18) and 48.8 per cent of men (Table 20) in the
sample reported having experienced this form of violence. Other forms of psychological
violence affected only small numbers of women and men in Ghana over their lifetime and
in the 12 months prior to the survey (Tables 19–21).
Age patterns. The results show that age was less strongly correlated to lifetime experiences
of psychological violence than to the other types of violence discussed above, with no clear
pattern of association for either women or men (Tables 18 and 20). But, in line with
discussions above for other forms of violence, younger women and men (aged 15–19 years)
were more likely to have reported at least one incident of psychological violence in the 12
months prior to the survey than all other older age groups: 38.8 per cent and 38.9 per cent
of women and men, respectively, aged 15–19 years experienced psychological violence in
that period, in comparison to 14.8 per cent of women and 18.8 per cent of men aged 50–60
years (Tables 19 and 21).
Employment status. The incidence of lifetime experiences of psychological violence was
highest among self-employed women (49.5 per cent) and employed men (56.1 per cent). It
was lowest among women who were not working (44.7 per cent) and among self-employed
men (49.2 per cent) (Tables 18 and 20). These employment patterns were also observed
when considering the incidence of psychological violence in the 12 months prior to the
survey (Tables 19–21).
Marital status. Men and women who were divorced, separated or widowed were about 6
percentage points more likely to have experienced psychological violence over their
lifetime than all other marital statuses on average (Tables 18 and 20). The lifetime incidence
of psychological violence across marital status was similar to what was found for the other
forms of violence analysed so far, and when considering the 12 months prior to the survey
(Tables 19–21).
Geographical patterns. Women and men living in rural areas (50.3 and 52.8 per cent,
respectively) were more likely to have reported at least one type of psychological violence
over their lifetime than those living in urban areas (45.8 and 51 per cent, respectively). As
94
with other forms of violence, the incidence of psychological violence was highest in the
Eastern region (73.7 per cent for women and 70.3 per cent for men). The lowest incidence
of lifetime psychological violence was observed in the Upper East region for women (13.1
per cent) and in the Upper West region for men (23.9 per cent) (Tables 18 and 20). These
geographical patterns were similar when considering the incidence of psychological
violence in the 12 months prior to the survey (Tables 19–21).
Education levels and asset ownership. Women without formal education and women with
higher education were up to 6.8 percentage points less likely to report having experienced
psychological violence than women with intermediate levels of education (primary,
middle/junior secondary school (JSS)/junior high school (JHS), secondary and technical
education) (Table 18). The incidence of psychological violence among men steadily
increased with the level of education until secondary/technical education. Men with higher
levels of education were as likely as those with primary education to experience
psychological violence (Table 20). Women in the highest asset quintile were 7.3 percentage
points less likely to report having experienced psychological violence over their lifetime
than women in the lowest asset quintile, whereas the incidence of psychological violence
was highest among men in the highest asset quintile (54.5 per cent). These patterns were
similar when considering the 12 months prior to the survey (Tables 19–21).
95
Table 18: Percentage of women who have experienced domestic or non-domestic psychological violence over their lifetime
Insulted,
humiliated or
belittled
Ignored or
threatened to be
abandoned
Scared or
intimidated
on purpose
Threatened
with the use
of a weapon
Threatened to
be hurt
Any
psychological
violence
Age group (years)
15–19 45.9 11.3 3.3 1.8 0.3 49.3 20–24 46.0 10.1 4.3 1.0 0.8 47.9 25–29 48.2 8.5 4.7 2.1 0.8 50.1 30–39 42.4 7.8 4.3 3.6 1.8 45.0 40–49 48.4 8.5 6.6 2.6 1.1 50.5 50–60 47.0 7.0 7.6 1.9 2.4 47.6
Employment
Self-employed 47.5 8.4 5.5 2.6 1.7 49.5 Employed 42.6 9.8 6.6 3.6 0.6 46.2 Not working 43.0 7.9 3.8 1.4 0.9 44.7
Marital status
Never married 39.9 4.1 2.6 0.8 0.0 41.0 Married or living together 44.7 7.7 5.0 2.1 1.0 46.7 Divorced/separated/widowed 51.4 12.1 6.7 4.2 3.1 54.0
Residence
Urban 44.0 8.4 5.3 2.1 1.0 45.8 Rural 48.1 8.4 5.1 2.9 1.8 50.3
96
Insulted,
humiliated or
belittled
Ignored or
threatened to be
abandoned
Scared or
intimidated
on purpose
Threatened
with the use
of a weapon
Threatened to
be hurt
Any
psychological
violence
Region
Western 54.0 4.7 1.7 1.1 1.0 56.7 Central 45.6 16.3 3.7 2.1 1.8 50.2 Greater Accra 36.4 7.2 5.3 2.3 1.0 37.7 Volta 59.1 10.9 20.7 1.2 2.9 59.5 Eastern 71.1 12.3 7.0 5.6 3.0 73.7 Ashanti 47.6 8.2 1.1 2.6 0.9 48.9 Brong Ahafo 38.1 8.5 5.2 3.7 0.7 41.4 Northern 41.7 2.8 5.2 1.0 1.0 42.8 Upper East 12.0 0.0 1.1 0.0 1.1 13.1 Upper West 14.9 5.3 1.5 1.2 0.0 17.3
Education level
None 42.4 7.5 6.0 2.3 1.4 44.1 Primary 49.7 9.5 4.3 2.9 2.7 51.5 Middle/JSS/JHS 47.5 9.1 4.9 2.4 0.8 49.8 Secondary 43.5 7.5 4.3 1.0 0.8 45.4 Technical 49.5 6.1 10.5 8.2 4.0 49.5 Higher 42.0 7.8 7.1 2.4 1.1 45.1
Asset quintile
Lowest 51.8 10.2 8.8 3.2 2.4 53.5 Second 44.2 9.0 3.6 2.9 1.3 46.8 Middle 50.9 7.1 3.6 2.1 0.6 52.6 Fourth 40.7 6.3 3.8 1.0 1.1 42.3 Highest 43.7 9.8 7.4 3.2 1.6 46.2
Total 45.9 8.4 5.2 2.4 1.4 48.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
97
Table 19: Percentage of women who experienced domestic or non-domestic psychological violence in the last 12 months
Insulted,
humiliated
or belittled
Ignored or
threatened to be
abandoned
Scared or
intimidated
on purpose
Threatened
with the use
of a weapon
Threatened to
be hurt
Any
psychological
violence
Age group (years)
15–19 37.0 7.7 1.9 1.0 0.0 38.8 20–24 24.1 5.1 0.7 0.0 0.4 26.4 25–29 24.4 3.0 1.4 1.0 0.3 25.6 30–39 19.8 3.0 0.9 1.2 0.8 21.5 40–49 16.6 2.3 1.2 0.6 0.4 19.0 50–60 14.2 0.7 0.8 0.6 0.0 14.8
Employment
Self-employed 19.4 2.6 0.7 0.7 0.4 20.9 Employed 18.8 3.2 2.3 1.4 0.0 21.7 Not working 25.8 4.4 1.5 0.9 0.5 27.4
Marital status
Never married 28.3 3.5 1.7 0.8 0.0 29.3 Married or living together 19.7 3.0 0.8 0.7 0.3 21.3 Divorced/separated/widowed 23.2 3.4 1.7 1.0 0.9 24.8
Residence
Urban 21.1 3.2 1.2 0.6 0.3 22.4 Rural 20.7 2.9 0.8 1.0 0.6 22.6
98
Insulted,
humiliated
or belittled
Ignored or
threatened to be
abandoned
Scared or
intimidated
on purpose
Threatened
with the use
of a weapon
Threatened to
be hurt
Any
psychological
violence
Region
Western 25.8 0.9 0.4 0.4 0.3 27.0 Central 22.0 7.2 0.9 1.1 0.7 25.5 Greater Accra 21.3 3.0 2.0 0.8 0.2 22.3 Volta 22.7 3.4 2.4 0.4 0.0 24.3 Eastern 28.7 4.0 2.5 2.6 1.2 32.1 Ashanti 21.0 3.1 0.2 0.6 0.3 22.7 Brong Ahafo 15.5 3.0 0.3 0.7 0.7 16.8 Northern 16.4 0.7 0.3 0.0 0.3 16.4 Upper East 4.9 0.0 0.0 0.0 0.0 4.9 Upper West 11.1 1.7 0.0 0.0 0.0 12.3
Education level
None 16.6 2.1 1.0 0.3 0.4 18.0 Primary 21.6 4.2 1.1 1.4 1.0 24.3 Middle/JSS/JHS 23.6 3.1 0.8 0.7 0.2 24.8 Secondary 22.6 4.1 1.8 0.6 0.3 24.7 Technical 16.4 2.0 0.9 4.8 1.4 17.6 Higher 16.6 2.0 1.5 0.6 0.2 18.4
Asset quintile
Lowest 22.1 3.1 1.6 1.2 0.7 25.0 Second 19.0 3.3 0.5 0.8 0.6 20.4 Middle 26.8 3.8 0.9 0.6 0.3 27.9 Fourth 20.5 2.2 1.4 0.2 0.2 21.3 Highest 17.1 3.3 0.9 1.2 0.2 19.0
Total 20.9 3.1 1.1 0.8 0.4 22.5
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
99
Table 20: Percentage of men who have experienced domestic or non-domestic psychological violence over their lifetime
Insulted,
humiliated
or belittled
Ignored or
threatened to be
abandoned
Scared or
intimidated
on purpose
Threatened
with the use
of a weapon
Threatened to
be hurt
Any
psychological
violence
Age group (years)
15–19 50.2 6.8 8.2 2.0 0.8 55.6 20–24 50.6 7.5 6.1 3.2 2.5 53.7 25–29 47.4 4.8 5.9 1.7 3.2 51.0 30–39 45.9 6.8 9.7 6.9 2.7 49.6 40–49 48.3 6.2 6.9 3.2 1.9 49.5 50–60 53.0 7.6 9.1 2.6 2.5 55.4
Employment
Self-employed 46.6 5.6 7.4 3.5 2.3 49.2 Employed 53.1 7.7 8.5 4.2 3.1 56.1 Not working 48.7 7.4 7.4 3.1 1.8 53.5
Marital status
Never married 40.7 6.4 6.3 2.9 1.1 44.4 Married or living together 49.5 6.4 7.5 3.7 2.3 52.2 Divorced/separated/widowed 53.9 7.5 9.0 3.1 3.5 58.1
Residence
Urban 47.5 6.5 7.4 3.8 3.1 51.0 Rural 50.0 6.6 8.0 3.4 1.8 52.8
100
Insulted,
humiliated
or belittled
Ignored or
threatened to be
abandoned
Scared or
intimidated
on purpose
Threatened
with the use of
a weapon
Threatened
to be hurt
Any
psychological
violence
Region
Western 57.8 6.6 7.6 5.7 2.7 61.9 Central 47.1 21.8 9.0 7.1 4.8 53.5 Greater Accra 41.1 3.8 5.9 2.9 2.9 44.6 Volta 57.1 7.8 23.9 1.5 0.0 59.2 Eastern 67.9 7.8 5.2 3.5 3.9 70.3 Ashanti 53.0 5.5 4.5 5.3 3.5 56.5 Brong Ahafo 41.3 6.9 9.4 3.8 2.4 45.2 Northern 42.2 3.1 4.8 0.2 0.0 43.1 Upper East 22.8 1.7 1.7 1.2 0.0 24.6 Upper West 23.9 0.0 0.0 1.5 0.0 23.9
Education level
None 39.4 3.9 5.7 1.2 0.3 40.9 Primary 49.4 7.7 8.2 2.9 2.8 53.0 Middle/JSS/JHS 47.7 5.7 6.5 4.0 2.1 51.2 Secondary 53.6 8.6 10.0 4.2 3.7 56.5 Technical 63.7 7.5 11.0 3.8 0.0 70.1 Higher 49.6 6.9 8.3 4.7 3.5 52.0
Asset quintile
Lowest 47.1 5.5 10.7 2.3 0.2 49.7 Second 47.9 8.1 6.7 3.3 2.6 50.9 Middle 50.9 7.1 8.3 3.7 4.4 54.3 Fourth 47.7 6.6 5.9 3.9 2.0 51.2 Highest 51.2 4.8 7.0 5.3 3.3 54.5
Total 48.8 6.5 7.7 3.6 2.4 51.9
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
101
Table 21: Percentage of men who experienced domestic or non-domestic psychological violence in the last 12 months
Insulted,
humiliated
or belittled
Ignored or
threatened to be
abandoned
Scared or
intimidated
on purpose
Threatened
with the use
of a weapon
Threatened to
be hurt
Any
psychological
violence
Age group (years)
15–19 36.1 4.0 2.4 1.2 0.8 38.9 20–24 34.0 4.3 2.0 1.5 1.5 38.2 25–29 28.0 1.5 2.5 0.5 1.7 29.5 30–39 22.4 2.9 3.2 1.7 1.1 25.2 40–49 20.9 2.0 0.8 0.6 0.5 22.5 50–60 17.9 1.8 0.6 0.3 0.0 18.8
Employment
Self-employed 21.3 2.0 1.4 1.0 0.6 23.0 Employed 30.4 2.9 3.3 0.9 1.8 33.0 Not working 30.7 3.9 2.0 1.2 0.8 34.1
Marital status
Never married 25.6 3.6 1.2 1.8 0.2 27.9 Married or living together 24.0 2.5 2.1 0.9 1.0 26.4 Divorced/separated/widowed 32.3 2.5 2.5 0.5 1.6 34.3
Residence
Urban 26.1 2.4 1.8 1.2 1.5 28.7 Rural 25.4 2.9 2.2 0.8 0.5 27.4
102
Insulted,
humiliated
or belittled
Ignored or
threatened to be
abandoned
Scared or
intimidated
on purpose
Threatened
with the use
of a weapon
Threatened to
be hurt
Any
psychological
violence
Region
Western 31.0 1.3 1.3 0.7 0.4 32.3 Central 28.2 8.3 5.1 1.9 2.3 36.4 Greater Accra 26.0 2.2 3.0 1.4 1.9 28.2 Volta 24.4 4.6 1.5 0.5 0.0 26.5 Eastern 37.7 3.4 1.7 0.9 2.1 41.5 Ashanti 30.0 2.3 1.4 1.1 0.6 31.7 Brong Ahafo 17.4 1.1 2.7 2.0 1.4 18.8 Northern 17.7 1.1 1.2 0.0 0.0 18.4 Upper East 10.2 1.7 1.1 0.0 0.0 11.9 Upper West 6.8 0.0 0.0 0.0 0.0 6.8
Education level
None 16.8 0.8 0.3 0.3 0.3 17.2 Primary 25.3 3.4 3.1 0.3 1.2 28.9 Middle/JSS/JHS 26.0 2.1 1.4 1.5 0.6 28.1 Secondary 31.0 4.8 2.8 1.3 1.1 33.0 Technical 26.4 6.3 0.6 0.0 0.0 32.6 Higher 26.5 1.4 3.6 0.9 2.5 29.4
Asset quintile
Lowest 22.2 2.1 0.9 0.2 0.2 23.3 Second 26.7 3.6 2.7 1.2 0.7 29.7 Middle 24.9 3.1 1.0 0.0 1.2 27.3 Fourth 28.2 2.8 2.5 2.0 1.1 30.4 Highest 26.5 1.3 2.9 1.4 1.9 29.2
Total 25.7 2.7 2.0 1.0 1.0 28.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
103
4.1.5. Economic violence
About one quarter of respondents – 28.2 and 25.8 per cent of women and men, respectively
– experienced an act of economic violence at least once during their lifetime; about one in
eight men and women experienced at least one type of economic violence in the 12 months
preceding the survey (Table 5). Tables 22–25 provide disaggregated information on the
incidence of economic violence across different population groups and socio-economic
characteristics, which is discussed in the paragraphs below.
Dimensions of economic violence. The most common form of economic violence
experienced by both women and men over their lifetime was the denial of chop money: 20.2
per cent of women and 12.2 per cent of men reported this form of violence (Tables 22 and
24). This was also the most prevalent form of violence reported by women and men in the
12 months prior to the survey (11.6 and 5.7 per cent, respectively) (Tables 23 and 25).
Age patterns. Younger age is again a risk factor for lifetime economic violence, particularly
among men. Around 30 per cent of men aged 15–29 years reported having experienced
economic violence in their lifetime, compared to 23.9 per cent of men aged 30–39 years,
21.6 per cent of men aged 40–49 years and 20.1 per cent of men aged 50–60 years (Table
24). The association between age and incidence of economic violence was less pronounced
among women. The highest incidence was among women aged 25–29 years (33.9 per cent),
while the lowest was among women aged 50–60 years (23.6 per cent) (Table 22). These age
patterns were also observed when considering the incidence of economic violence in the
12 months prior to the survey (Table 23–25).
Employment status. Just over 30 per cent of women who were not working at the time of
the survey experienced economic violence, compared to 28.3 per cent of self-employed
women and 23.1 per cent of employed women (Table 22). In contrast, economic violence
was more prevalent among employed men: 29.4 per cent of employed men reported having
experienced economic violence, compared to 28.3 per cent of men not working and 22.9
per cent of self-employed men (Table 24). These results suggest that employed men were
at higher risk of economic violence, perhaps because economic expectations may be higher
among these men. On the other hand, economic violence was more prevalent among
women who stay at home, perhaps as a result of their economic dependence on men, as
discussed in the literature surveyed in Section 2. The same patterns were observed when
the survey questions referred to the 12 months prior to the survey (Tables 23–25).
Marital status. As with other forms of violence, lifetime experiences of economic violence
were also more prominent among men and women who were divorced, separated or
widowed at the time of the survey. Thirty-three per cent of divorced, separated or
widowed women experienced economic violence, compared to 18.6 per cent of never-
married women and 27.6 per cent of married women or women living with a partner (Table
22). Almost 35 per cent of divorced, separated or widowed men experienced economic
violence, compared to 25.2 per cent of never-married men and 23.4 per cent of men married
104
or living with a partner (Table 24). These patterns were similar when considering the 12
months prior to the survey (Tables 23–25).
Geographical patterns. The incidence of economic violence was higher among women
living in rural areas (31.3 per cent) than women living in urban areas (25.5 per cent) (Table
22), whereas there was almost no difference in the lifetime incidence of economic violence
among men living in urban or rural areas (Table 24). As with other forms of violence, both
women and men living in the Eastern region (48.7 per cent and 51.9 per cent, respectively)
were more likely than individuals in other regions to have experienced at least one form
of economic violence over their lifetime. This distribution of the incidence of economic
violence across locations was similar for the 12 months prior to the survey (Tables 23–25).
Education levels and asset ownership. The incidence of economic violence was lower
among women with higher education (19.4 per cent) than among women with no formal
education or intermediate levels of education (primary, middle/JSS/JHS, secondary,
technical). Women with primary education were more likely than women with other
education levels to have experienced economic violence (35.8 per cent) (Table 22). In
contrast, the incidence of economic violence was lowest among men without formal
education (10.9 per cent). Wealthier women in the two highest asset quintiles were around
5–10 percentage points less likely to have experienced economic violence over the course
of their lives. In contrast, men in the four highest asset quintiles were more likely to
experience economic violence (26.7, 27.3, 27.1 and 26 per cent, respectively) than men in
the lowest asset quintile (22 per cent) (Table 24). These education and wealth patterns were
similar when considering the 12 months prior to the survey (Tables 23–25).
105
Table 22: Percentage of women who have experienced domestic or non-domestic economic violence over their lifetime
Denied
money
Cash
taken
out
Belongings
controlled
Property
damaged
Prohibited
from
working
Forced
to work
Denied
food
Any
economic
violence
Age group (years)
15–19 19.4 1.4 3.4 1.6 2.5 6.8 1.1 30.4 20–24 21.3 2.5 1.9 3.6 2.4 6.2 0.1 31.7 25–29 25.9 3.6 2.6 5.2 3.9 4.7 0.0 33.9 30–39 20.4 2.6 2.1 2.6 3.0 3.9 0.2 27.4 40–49 15.8 2.2 1.8 3.2 2.7 2.6 0.1 24.2 50–60 16.8 2.3 1.8 3.6 1.9 3.6 0.1 23.6
Employment
Self-employed 20.6 2.8 1.7 3.0 3.0 4.2 0.2 28.3 Employed 15.6 1.8 1.8 4.1 1.5 2.9 0.0 23.1 Not working 21.1 2.3 3.5 4.1 3.0 5.0 0.1 30.2
Marital status
Never married 11.5 0.8 1.9 0.8 0.3 4.3 0.1 18.6 Married or living together 19.4 2.8 2.1 3.3 2.8 4.4 0.1 27.6 Divorced/separated/widowed 24.9 2.4 2.4 4.3 3.6 3.7 0.4 33.0
Residence
Urban 16.9 2.0 2.6 4.1 2.7 4.3 0.2 25.5 Rural 23.8 3.2 1.7 2.7 3.1 4.3 0.2 31.3
106
Denied
money
Cash
taken
out
Belongings
controlled
Property
damaged
Prohibited
from
working
Forced
to work
Denied
food
Any
economic
violence
Region
Western 31.8 2.3 2.1 4.4 2.8 1.5 0.7 37.4 Central 25.1 2.2 3.3 4.0 2.6 2.7 0.1 33.8 Greater Accra 11.4 0.4 2.1 4.2 2.1 4.6 0.0 19.1 Volta 13.0 1.4 2.1 0.4 0.2 2.0 0.1 17.4 Eastern 33.2 13.1 2.6 6.8 4.9 7.2 0.2 48.7 Ashanti 20.2 2.3 2.0 2.8 2.2 2.7 0.3 27.3 Brong Ahafo 24.3 1.6 1.9 2.3 5.4 8.7 0.1 34.2 Northern 12.4 0.3 1.1 2.8 4.0 7.2 0.0 20.8 Upper East 6.7 0.0 1.7 1.2 0.0 0.6 0.0 9.6 Upper West 16.6 0.0 2.4 1.2 4.8 2.6 0.0 20.4
Education level
None 16.6 1.4 2.2 3.6 3.2 3.5 0.1 24.2 Primary 24.1 4.4 2.6 4.0 2.8 5.8 0.1 35.8 Middle/JSS/JHS 23.5 3.1 1.9 3.2 2.8 4.7 0.3 30.8 Secondary 15.7 1.2 2.0 2.4 2.8 4.0 0.0 22.0 Technical 16.9 2.9 1.1 1.2 5.4 1.4 0.0 25.6 Higher 11.8 1.7 2.4 5.2 0.4 1.9 0.0 19.4
Asset quintile
Lowest 21.5 1.8 1.9 2.8 3.8 5.1 0.3 30.7 Second 23.1 3.1 2.6 4.1 2.5 3.3 0.1 30.9 Middle 23.8 1.8 2.7 3.2 2.6 4.5 0.5 31.7 Fourth 17.3 3.4 1.6 3.2 2.5 4.9 0.0 25.5 Highest 13.4 2.3 1.9 3.6 3.1 3.8 0.0 20.6
Total 20.2 2.6 2.2 3.4 2.8 4.3 0.2 28.2
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
107
Table 23: Percentage of women who experienced domestic or non-domestic economic violence in the last 12 months
Denied
money
Cash
taken
out
Belongings
controlled
Property
damaged
Prohibited
from
working
Forced to
work
Denied
food
Any
economic
violence
Age group (years)
15–19 15.0 1.4 1.6 0.9 0.4 2.8 5.5 20.4 20–24 13.8 1.1 1.0 1.9 1.1 2.2 1.9 17.9 25–29 17.1 1.5 1.0 2.0 1.9 2.0 1.3 20.8 30–39 12.6 1.2 1.1 0.9 1.4 0.3 1.3 15.4 40–49 7.9 0.5 0.3 0.0 0.9 0.5 0.2 8.9 50–60 4.2 0.3 0.1 0.3 0.5 0.0 0.3 5.6
Employment
Self-employed 11.4 1.1 0.5 0.6 1.4 0.8 1.1 13.7 Employed 7.6 1.3 1.0 1.2 1.0 0.7 0.3 11.0 Not working 14.2 0.7 1.5 1.7 0.7 1.7 2.4 17.8
Marital status
Never married 6.9 0.8 0.9 0.3 0.0 3.3 3.7 12.4 Married or living together 12.4 1.1 0.8 1.0 1.3 0.9 0.9 15.0 Divorced/separated/widowed 10.2 0.6 0.7 1.1 1.1 0.9 1.9 13.0
Residence
Urban 9.3 0.6 1.0 1.2 1.0 0.7 0.7 12.1 Rural 14.2 1.5 0.5 0.7 1.5 1.4 1.9 17.0
108
Denied
money
Cash
taken
out
Belongings
controlled
Property
damaged
Prohibited
from
working
Forced
to work
Denied
food
Any
economic
violence
Region
Western 22.9 0.5 0.4 2.8 0.3 0.0 1.2 23.3 Central 13.6 0.9 1.5 1.1 0.7 1.1 2.6 15.6 Greater Accra 7.2 0.0 1.1 0.8 0.7 1.0 0.3 9.8 Volta 5.7 0.5 0.0 0.0 0.2 1.3 0.7 6.9 Eastern 16.1 5.4 1.1 0.8 1.2 1.1 1.3 23.4 Ashanti 12.4 1.3 0.4 1.0 1.3 0.5 1.4 15.3 Brong Ahafo 11.7 0.0 0.5 0.8 3.6 1.6 1.9 15.0 Northern 6.9 0.3 0.9 0.9 1.8 2.1 2.5 9.4 Upper East 6.1 0.0 1.1 0.6 0.0 0.6 0.6 7.9 Upper West 12.5 0.0 2.4 0.0 2.4 1.5 0.0 13.9
Education level
None 8.6 0.3 0.7 1.3 1.1 0.5 1.0 10.2 Primary 15.0 2.2 0.9 0.1 0.9 1.0 0.8 18.6 Middle/JSS/JHS 13.1 1.2 0.6 1.2 1.3 1.3 2.0 16.2 Secondary 10.8 0.4 0.8 1.2 1.5 1.3 0.9 13.9 Technical 9.8 0.0 1.1 0.0 3.3 0.0 1.9 12.1 Higher 6.7 0.7 1.4 0.3 0.0 0.7 0.0 9.3
Asset quintile
Lowest 11.2 1.0 0.5 0.8 1.5 1.6 2.2 14.6 Second 14.2 1.5 0.8 1.5 1.4 1.1 1.3 17.0 Middle 15.7 0.4 1.1 0.8 0.7 0.6 2.0 17.3 Fourth 9.7 1.0 1.0 0.8 1.1 1.3 0.5 13.0 Highest 6.2 0.8 0.7 0.6 1.1 0.1 0.5 8.6
Total 11.6 1.0 0.8 1.0 1.2 1.0 1.3 14.4
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
109
Table 24: Percentage of men who have experienced domestic or non-domestic economic violence over their lifetime
Denied
money
Cash
taken out
Belongings
controlled
Property
damaged
Prohibited
from
working
Forced to
work
Denied
food
Any
economic
violence
Age group (years)
15–19 21.4 4.7 4.1 4.0 1.3 6.6 0.1 32.3 20–24 17.2 4.2 0.6 6.4 4.5 7.4 0.1 29.2 25–29 13.4 4.5 0.4 5.8 3.1 9.6 0.1 30.0 30–39 8.7 3.8 1.4 9.1 3.8 6.2 0.3 23.9 40–49 9.2 3.7 2.2 7.2 3.7 2.9 0.7 21.6 50–60 8.2 4.2 1.8 6.2 0.8 4.2 0.4 20.1
Employment
Self-employed 9.7 4.3 0.9 6.8 3.3 5.3 0.3 22.9 Employed 12.2 5.7 2.5 6.5 3.4 9.5 0.4 29.4 Not working 18.3 1.8 2.0 6.9 2.2 4.5 0.3 28.3
Marital status
Never married 16.1 4.5 2.3 4.9 1.1 5.0 0.1 25.2 Married or living together 9.2 3.9 1.5 6.4 3.0 6.4 0.3 23.4 Divorced/separated/widowed 18.7 4.7 1.3 9.3 5.1 6.7 0.1 34.5
Residence
Urban 12.4 3.8 2.1 7.8 2.9 6.9 0.3 25.9 Rural 12.0 4.4 1.1 5.8 3.2 5.6 0.3 25.7
110
Denied
money
Cash
taken
out
Belongings
controlled
Property
damaged
Prohibited
from
working
Forced
to work
Denied
food
Any
economic
violence
Region
Western 18.9 4.0 1.4 6.8 2.5 6.5 0.0 32.8 Central 23.1 6.0 6.4 12.7 3.1 4.9 1.1 40.6 Greater Accra 12.0 0.3 2.1 9.2 2.4 8.9 0.0 23.0 Volta 4.5 2.4 1.6 1.0 0.9 4.9 0.1 13.4 Eastern 23.0 21.1 3.1 10.5 8.1 5.7 0.2 51.9 Ashanti 11.8 4.9 0.2 8.1 2.5 4.3 0.8 25.6 Brong Ahafo 7.4 0.5 0.9 5.5 5.3 7.9 0.0 22.5 Northern 3.9 0.0 0.0 3.2 3.0 8.2 0.4 15.1 Upper East 4.0 0.0 0.0 0.0 0.0 1.6 0.0 5.7 Upper West 6.6 0.0 0.0 0.0 2.2 6.5 0.1 11.0
Education level
None 3.4 0.6 0.0 1.6 1.2 5.7 0.0 10.9 Primary 13.9 2.8 2.8 6.1 2.8 6.8 0.1 26.8 Middle/JSS/JHS 13.7 7.1 1.9 8.2 3.3 5.0 0.6 29.4 Secondary 15.6 3.2 1.0 7.1 4.6 8.5 0.1 28.6 Technical 11.0 3.4 1.5 9.3 1.5 8.2 0.1 28.7 Higher 11.1 2.8 2.0 7.2 2.7 5.9 0.3 25.3
Asset quintile
Lowest 11.6 2.2 1.7 4.8 2.1 7.1 0.4 22.0 Second 12.5 5.2 1.3 5.0 4.3 6.3 0.2 26.7 Middle 12.7 3.8 1.8 8.3 2.5 6.6 0.1 27.3 Fourth 13.4 4.2 1.8 9.0 3.0 4.4 0.1 27.1 Highest 10.5 5.0 1.3 7.8 3.0 6.7 0.9 26.0
Total 12.2 4.1 1.6 6.8 3.1 6.2 0.3 25.8
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
111
Table 25: Percentage of men who experienced domestic or non-domestic economic violence in the last 12 months
Denied
money
Cash
taken
out
Belongings
controlled
Property
damaged
Prohibited
from
working
Forced to
work
Denied
food
Any
economic
violence
Age group (years)
15–19 15.1 3.4 3.5 3.1 0.5 3.8 2.9 23.3 20–24 9.5 3.0 0.1 3.8 3.1 2.7 3.2 18.9 25–29 5.5 1.5 0.0 2.2 1.4 5.0 0.9 14.2 30–39 2.2 0.8 0.4 3.2 1.2 1.1 0.3 8.3 40–49 3.9 0.5 0.6 2.2 0.9 0.5 1.0 8.5 50–60 3.4 0.9 0.8 1.2 0.3 0.7 0.6 5.9
Employment
Self-employed 3.5 1.4 0.1 1.8 1.1 1.0 0.5 8.1 Employed 4.5 1.6 0.9 2.7 1.8 4.8 2.2 14.2 Not working 12.8 1.6 2.0 4.5 1.1 2.1 2.2 20.0
Marital status
Never married 11.0 2.7 1.9 3.1 0.4 2.9 2.9 18.3 Married or living together 3.8 1.0 0.5 1.8 1.0 1.9 0.7 9.4 Divorced/separated/widowed 7.1 2.1 0.3 4.4 3.1 2.6 2.0 16.4
Residence
Urban 5.5 1.4 1.2 3.1 1.2 2.9 1.2 13.3 Rural 6.0 1.6 0.2 2.1 1.3 1.5 1.4 11.3
112
Denied
money
Cash
taken
out
Belongings
controlled
Property
damaged
Prohibited
from
working
Forced
to work
Denied
food
Any
economic
violence
Region
Western 9.3 0.4 0.3 1.2 0.0 2.4 0.5 13.9 Central 12.4 3.8 4.5 4.3 1.3 1.9 9.1 25.7 Greater Accra 6.3 0.0 1.4 4.0 1.3 4.3 0.6 13.0 Volta 2.0 1.7 0.0 0.4 0.9 0.2 0.8 4.6 Eastern 3.1 9.0 0.8 2.6 4.8 2.3 2.1 19.2 Ashanti 7.2 0.6 0.0 4.8 0.9 1.6 0.7 13.4 Brong Ahafo 4.8 0.5 0.0 1.5 1.9 0.9 0.4 9.2 Northern 2.2 0.0 0.0 1.4 0.2 3.0 0.3 6.5 Upper East 2.4 0.0 0.0 0.0 0.0 1.0 0.0 3.4 Upper West 2.2 0.0 0.0 0.0 2.2 3.8 0.0 3.8
Education level
None 0.6 0.0 0.0 1.1 0.0 0.6 0.3 2.3 Primary 7.3 0.9 1.0 0.3 2.1 2.4 3.5 12.5 Middle/JSS/JHS 6.2 2.4 1.1 3.1 1.3 1.6 1.2 13.3 Secondary 8.8 2.0 0.1 3.5 2.0 2.9 1.3 16.7 Technical 7.8 0.0 0.0 3.2 1.5 3.8 1.5 15.2 Higher 3.3 0.6 1.1 3.6 0.7 3.8 0.6 12.0
Asset quintile
Lowest 7.1 0.5 0.8 1.7 0.8 2.4 1.1 9.4 Second 5.6 1.6 0.5 2.5 2.0 2.1 2.5 13.2 Middle 5.6 1.9 0.3 1.4 1.4 2.0 1.2 11.7 Fourth 6.5 2.1 1.0 3.4 1.0 1.4 0.6 13.7 Highest 3.5 1.3 0.9 4.3 1.0 3.3 0.6 13.3
Total 5.7 1.5 0.7 2.6 1.3 2.2 1.3 12.3
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
113
4.2. Incidence of domestic violence against women and men in Ghana
This subsection and the remainder of the report will focus on violence experienced or
perpetrated within domestic relations – domestic violence – in the 12 months prior to the
survey.24 Domestic violence is defined, as per the Ghana Domestic Violence Act 732, as any
form of violence against men or women perpetrated by a domestic relation. Domestic
relations are core members of the family, as well as non-family members who live within
the household. Table 26 summarises the overall incidence of domestic violence across the
five types of violence in the 12 months prior to the survey.
The most common form of domestic violence reported by women in the 12 months prior
to the GFLHS 2015 was economic violence (12.8 per cent), followed by social violence (11.6
per cent) and psychological violence (9.3 per cent) (Table 26). This order is different from
the previous section, where psychological violence had the highest incidence (Table 5).
Domestic violence among men was mainly psychological violence (7.9 per cent), followed
by social violence (7.7 per cent) and economic violence (7.3 per cent) (Table 26). This order
is the same as for the overall levels of violence discussed in the previous section (Table 5).
Table 26: Incidence of domestic violence in Ghana 2015 in the 12 months prior to the
survey (in percentages)
Social
violence
Physical
violence
Sexual
violence
Psychological
violence
Economic
violence
Total
(at least
one type)
Women 11.6 6.0 2.5 9.3 12.8 27.7
Men 7.7 2.1 1.4 7.9 7.3 20.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
It is important to note that these types of domestic violence did not occur in isolation.
According to the GFLHS 2015, many respondents who reported having experienced
domestic violence experienced multiple forms and types of violence: 23.3 per cent of
women and 18.9 per cent of men who experienced domestic violence reported having
experienced two types of domestic violence, while 9.4 per cent of women and 5.7 per cent
of men reported having experienced three different types of domestic violence. In what
follows, the analysis focuses on each type of violence for ease of discussion.
24 Qualitative and quantitative data on domestic violence covered only the 12 months prior to the survey. No
questions were asked about lifetime experiences due to budget and time constraints, as this would have
increased substantially the length of the questionnaires and interviews.
114
4.2.1. Domestic social violence
Table 27 compares the proportion of domestic and non-domestic social violence. The
results show that almost all types of social violence against men and women were largely
perpetrated within domestic relations. The only exception was the spread of rumours,
photos and videos without permission: only 23.4 per cent and 13.2 per cent of this form of
violence against women and men, respectively, constituted domestic violence. Nearly all
(97.4 per cent) of the instances reported by women of being prevented from seeing friends
or family were perpetrated by domestic relations. The following forms of violence against
women were committed mostly by domestic relations: not being allowed to leave their
home (98.5 per cent), required to report on their activities (88.4 per cent), being stalked
(62.5 per cent) and being forced to have an abortion (women only) (84.6 per cent). The
results for men were similar, with the exception of stalking, which was mostly (59 per cent)
perpetrated by non-domestic relations.
Table 27: Percentage of domestic social violence over the last 12 months
Not
seeing
friends
or
family
Not
allowed
to leave
home
Required
to
report
activities
Stalked Rumour,
photos,
videos
spread
Forced to
have an
abortion
Women Perpetrated by family members 63.1 65.4 48.8 31.2 2.4 11.5
Perpetrated by non-family
members but domestic
34.3
33.1
40.9
32.8
21.2
73.1
% Domestic violence 97.4 98.5 88.4 62.5 23.4 84.6
% Non-domestic violence 2.6 1.5 11.6 37.5 76.6 15.4
Total 100.0 100.0 100.0 100.0 100.0 100.0
Men Perpetrated by family members 65.3 80.0 29.8 17.9 3.3 N/A
Perpetrated by non-family
members but domestic
27.1
15.0
43.3
23.1
10.3
N/A
% Domestic violence 92.4 95.0 73.1 41.0 13.2 N/A
% Non-domestic violence 7.6 5.0 26.9 59.0 86.8 N/A
Total 100.0 100.0 100.0 100.0 100.0 N/A
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Tables 28 and 29 illustrate how domestic social violence was distributed across the socio-
economic characteristics and geographical location of women and men, respectively. The
incidence of domestic social violence was higher among women (11.6 per cent) than among
men (7.7 per cent), and all individual acts of domestic social violence were more prominent
among women than among men. The relative importance of each type of domestic social
violence was similar among women and men. The most common form of domestic social
violence against women and men was being required to report their activities (5.1 per cent
115
for women and 3.9 per cent for men), followed by being stopped from leaving home (4.1
per cent for women and 1.6 per cent for men), being kept from seeing friends and family
(4.0 per cent for women and 1.4 per cent for men), being the victim of rumour, photo or
video spread (3.4 per cent for women and 1.7 per cent for men), and being stalked (1.2 per
cent for women and 1.0 per cent for men). In addition, 0.7 per cent of women in the GFLHS
2015 were forced to having an abortion.
Age patterns. As observed with overall levels of social violence, the incidence of domestic
social violence was highest among women in the youngest age group (15–19 years): 35.7
per cent of women in this age group experienced at least one type of social violence in the
12 months prior to the survey. The incidence in other age groups was considerably lower:
19.3 per cent for women aged 20–24 years, 15.3 per cent for women aged 25–29 years, 8.6
per cent for women aged 30–39 years, 5.0 per cent for women aged 40–49 years and 3.9 per
cent for women aged 50–60 years. The incidence of all acts of domestic social violence was
highest among women aged 15–19 years, with the exception of being stalked, which was
highest (2.4 per cent) for women aged 20–24 years. These age patterns were similar for
men, but differences across the age groups were less pronounced: 13.5 per cent of all men
aged 15–19 years experienced at least one form of domestic social violence in the 12 months
prior to the survey, in comparison to 13 per cent of men aged 20–24 years, 8.7 per cent of
men aged 25–29 years, 5.7 per cent of men aged 30–39 years, 3.8 per cent of men aged 40–
49 and 4.4 per cent of men aged 50–60 years.
Employment status. The incidence of domestic social violence was highest for women and
men who were not working at the time of the survey (19.8 per cent for women and 10.9 per
cent for men). Both women and men who were self-employed were at lower risk of
experiencing any act of domestic social violence (8.3 and 5.8 per cent, respectively). All
types of domestic social violence were more commonly reported by women and men who
were not working. The only exceptions were men who were required to report their
activities and men who were victims of a rumour, photo or video being spread. These acts
of domestic social violence were highest among employed men (5.6 per cent and 2 per cent,
respectively).
Marital status. Women who had never been married were at the highest risk of
experiencing any form of domestic social violence: 27.1 per cent of these women
experienced at least one act of domestic social violence in the 12 months prior to the survey,
in comparison with 11 per cent of married women and 9.9 per cent of women who were
divorced, separated or widowed. However, being a victim of the spreading of a rumour,
photo or video or being forced to having an abortion were more common among women
who were divorced, separated or widowed (4.6 per cent and 0.9 per cent, respectively). The
patterns for men were different. Men who were divorced, separated or widowed were
more at risk of experiencing any form of domestic social violence (8.3 per cent, in
comparison to 8.1 per cent for never-married men and 7.4 per cent for married men).
However, never-married men were more likely to report being stopped from leaving home
(3.8 per cent) and being kept from seeing friends and family (2.4 per cent). These latter
116
results are likely to be explained by the fact that these men lived at home with their parents,
an issue that will be explored in the next section, which focuses on the perpetration of
violence.
Geographical patterns. There was little difference in the levels of domestic social violence
experienced by women living in urban and rural areas (11.9 and 11.3 per cent,
respectively). The differences were more pronounced for men: 8.5 per cent of men living
in urban areas experienced at least one act of domestic social violence, in comparison to 6.9
per cent of men living in rural areas. Women and men living in the Western region were
more likely to report having experienced at least one act of domestic social violence (14.8
and 12.2 per cent, respectively). The incidence of domestic social violence was lowest
among women living in the Brong Ahafo region (7.4 per cent) and men living in the Upper
East region (1.7 per cent).
Education levels and asset ownership. Table 29 shows a linear association between
domestic social violence and the education and wealth status of men: the higher their level
of education and asset quintile, the higher the likelihood of men reporting having
experienced at least one act of domestic social violence in the 12 months prior to the survey.
Thus 10.4 per cent of men with higher education reported having experienced domestic
social violence, in comparison to 2.6 per cent of men with no education. Similarly, 11.9 per
cent of men in the highest asset quintile experienced domestic social violence, in
comparison to 4.0 per cent of men in the lowest asset quintile. As discussed before, this
result may be explained by the fact that better educated and wealthier men may also be
more likely to report experiences of domestic violence. The patterns for women were less
pronounced. Women with secondary levels of education (15.8 per cent) and in the middle
asset quintile (12.9 per cent) were the most likely to report having experienced domestic
social violence. The incidence of domestic social violence was lowest among women with
no education (5.4 per cent) and women in the highest asset quintile (10.3 per cent).
117
Table 28: Percentage of women who experienced domestic social violence over the last 12 months
Not
seeing
friends or
family
Not
allowed to
leave
home
Required
to
report
activities
Stalked Rumour,
photos,
videos
spread
Forced to
have an
abortion
Any
domestic
social
violence
Age group (years)
15–19 16.5 21.8 12.2 2.1 5.8 1.9 35.7 20–24 7.6 7.1 9.4 2.4 3.6 1.9 19.3 25–29 4.8 4.1 7.8 1.5 4.7 0.9 15.3 30–39 2.4 2.3 4.0 1.1 3.5 0.4 8.6 40–49 1.3 1.5 2.0 0.6 2.0 0.0 5.0 50–60 0.9 0.7 0.7 0.4 2.1 0.0 3.9
Employment
Self-employed 2.4 2.4 3.4 1.0 3.3 0.5 8.3 Employed 1.9 2.4 6.1 0.5 3.0 1.5 13.0 Not working 9.2 9.3 9.2 1.9 4.0 0.8 19.8
Marital status
Never married 13.1 14.2 8.9 1.6 3.5 0.5 27.1 Married or living together 3.6 3.8 5.5 1.2 3.1 0.6 11.0 Divorced/separated/widowed 3.1 2.7 2.7 1.3 4.6 0.9 9.9
Residence
Urban 4.0 3.5 5.7 1.1 3.4 0.7 11.9 Rural 4.0 4.7 4.3 1.3 3.4 0.7 11.3
118
Not
seeing
friends or
family
Not
allowed to
leave
home
Required
to
report
activities
Stalked Rumour,
photos,
videos
spread
Forced to
have an
abortion
Any
domestic
social
violence
Region
Western 4.7 6.4 3.9 0.6 4.6 2.0 14.8 Central 4.5 4.6 3.9 0.0 1.4 1.1 10.2 Greater Accra 3.0 3.1 5.5 1.9 4.0 0.1 11.1 Volta 5.4 6.2 6.1 1.9 4.6 0.4 12.7 Eastern 4.8 5.5 6.3 1.7 4.5 1.1 14.5 Ashanti 3.2 2.8 5.9 0.6 4.0 0.4 11.9 Brong Ahafo 1.8 1.4 5.0 1.9 1.8 1.0 7.4 Northern 7.2 4.9 2.9 1.4 0.9 0.2 10.9 Upper East 1.8 6.7 0.6 0.0 4.7 0.0 10.6 Upper West 8.0 5.6 9.0 0.0 2.2 0.0 12.5
Education level
None 2.3 2.2 1.6 0.6 1.6 0.2 5.4 Primary 4.0 4.9 4.9 1.5 6.0 0.4 13.4 Middle/JSS/JHS 5.0 5.3 5.1 1.5 4.3 1.1 13.7 Secondary 4.2 4.2 10.9 1.8 2.0 0.3 15.8 Technical 0.0 1.3 4.6 0.0 0.9 0.6 6.8 Higher 4.8 2.4 6.8 0.6 1.5 0.8 10.2
Asset quintile
Lowest 4.4 4.5 3.9 1.3 4.6 0.3 11.2 Second 4.2 4.5 4.3 1.6 2.8 0.9 11.1 Middle 3.5 4.6 4.9 1.6 4.2 1.6 12.9 Fourth 3.9 4.1 6.8 0.6 4.0 0.4 12.4 Highest 3.5 2.2 5.6 1.0 1.3 0.2 10.3
Total 4.0 4.1 5.1 1.2 3.4 0.7 11.6
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
119
Table 29: Percentage of men who experienced domestic social violence over the last 12 months
Not
seeing
friends or
family
Not
allowed to
leave
home
Required
to
report
activities
Stalked Rumour,
photos,
videos
spread
Forced to
have an
abortion
Any
domestic
social
violence
Age group (years)
15–19 4.6 6.8 3.3 0.0 3.4 N/A 13.5 20–24 2.2 1.0 6.3 1.7 2.9 N/A 13.0 25–29 1.0 1.4 6.0 2.2 1.3 N/A 8.7 30–39 0.5 0.5 4.0 0.7 1.4 N/A 5.7 40–49 0.8 0.7 1.5 0.7 1.1 N/A 3.8 50–60 0.8 1.8 1.7 0.0 1.1 N/A 4.4
Employment
Self-employed 1.2 1.4 2.9 1.0 1.5 N/A 5.8 Employed 0.8 0.7 5.6 0.8 2.0 N/A 8.6 Not working 2.5 3.3 4.2 1.1 1.9 N/A 10.9
Marital status
Never married 2.4 3.8 2.0 0.0 2.0 N/A 8.1 Married or living together 1.3 1.4 4.4 0.8 1.5 N/A 7.4 Divorced/separated/widowed 0.8 0.5 4.2 2.3 2.4 N/A 8.3
Residence
Urban 1.4 1.8 4.9 1.0 1.3 N/A 8.5 Rural 1.3 1.5 3.0 0.9 2.2 N/A 6.9
120
Not
seeing
friends or
family
Not
allowed to
leave
home
Required
to
report
activities
Stalked Rumour,
photos,
videos
spread
Forced to
have an
abortion
Any
domestic
social
violence
Region
Western 1.8 2.3 5.7 2.6 1.8 N/A 12.2 Central 2.3 1.6 4.7 1.8 5.9 N/A 10.0 Greater Accra 1.3 1.8 6.0 1.9 0.6 N/A 9.1 Volta 0.5 1.3 3.5 0.0 1.9 N/A 6.3 Eastern 1.8 2.1 4.5 0.6 2.7 N/A 9.6 Ashanti 1.1 1.0 4.7 0.6 1.8 N/A 8.5 Brong Ahafo 1.4 1.8 0.0 0.0 0.9 N/A 3.1 Northern 0.8 1.1 1.4 0.3 0.9 N/A 3.1 Upper East 0.7 0.7 1.0 0.0 0.0 N/A 1.7 Upper West 4.4 4.4 0.0 0.0 1.8 N/A 6.1
Education level
None 1.0 1.7 0.4 0.0 0.4 N/A 2.6 Primary 2.8 1.9 1.8 1.3 1.8 N/A 7.5 Middle/JSS/JHS 1.9 2.5 3.0 0.3 2.2 N/A 7.5 Secondary 0.5 0.9 4.9 2.2 2.5 N/A 9.3 Technical 2.6 0.0 6.8 1.8 0.6 N/A 9.4 Higher 0.3 0.5 9.2 1.1 0.9 N/A 10.4
Asset quintile
Lowest 1.1 1.5 1.2 0.4 1.4 N/A 4.0 Second 1.3 1.2 2.3 0.8 2.4 N/A 6.4 Middle 1.7 1.7 4.0 0.3 2.7 N/A 8.2 Fourth 1.9 2.1 4.5 1.0 1.3 N/A 9.2 Highest 0.8 1.8 9.0 2.5 0.7 N/A 11.9
Total 1.4 1.6 3.9 1.0 1.7 N/A 7.7
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
121
4.2.2. Domestic physical violence
All types of physical violence against women were mainly perpetrated by domestic
relations (within and outside the immediate family), whereas all types of physical violence
against men were mainly perpetrated by non-domestic relations (Table 30). Forms of
physical violence against women and men were more likely to be perpetrated by domestic
relations outside the immediate family. The exceptions were women being slapped or
being threatened with a weapon, which were equally likely to be perpetrated by family or
non-family members (36.8 and 38.9 per cent, respectively). It was also more likely that men
were burnt by family members (33.3 per cent) than by non-family members (none).
Table 30: Percentage of domestic physical violence over the last 12 months
Slapped or
thrown
things at
Pushed
or
shoved
Hit Kicked,
dragged
or beaten
up
Choked
or
strangled
Burnt Weapon
used
against
Women
Perpetrated by family
members
36.8 24.2 26.9 35.9 25.0 50.0 38.9
Perpetrated by non-
family members but
domestic
36.8
38.7
31.3
48.4
55.0
50.0
38.9
% Domestic violence 73.0 62.9 58.2 84.4 80.0 100.0 77.8
% Non-domestic
violence
27.0 37.1 41.8 15.6 20.0 0.0 22.2
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Men
Perpetrated by family
members
5.1 0.0 2.0 5.9 0.0 33.3 4.5
Perpetrated by non-
family members but
domestic
22.2
6.4
12.0
23.5
10.5
0.0
18.2
% Domestic violence 27.3 6.4 14.0 29.4 10.5 33.3 22.7 % Non-domestic
violence
72.7 93.6 86.0 70.6 89.5 66.7 77.3
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Tables 31 and 32 illustrate the variation of incidence of domestic physical violence across
the socio-economic and geographical characteristics of women and men, respectively.
Women were almost three times more likely to experience physical violence (6 per cent)
than men (2.1 per cent). The most common act of domestic physical violence experienced
by women and men was being slapped or having objects thrown at them (4.4 and 1.4 per
cent, respectively). The second most common form of domestic physical violence for
women was being kicked, dragged or beaten up (1.8 per cent), while for men it was being
hit (0.3 per cent). These results suggest that women and men experienced domestic
122
physical violence in different ways, an issue that will be further analysed in subsequent
sections.
Age patterns. As with other forms of domestic and non-domestic violence, the incidence of
domestic physical violence was highest among younger women (15–19 years old): 15.2 per
cent of these women reported having experienced at least one act of domestic physical
violence, in comparison to 9.4 per cent of women aged 20–24 years, 6.9 per cent of women
aged 25–29 years, 7.1 per cent of women aged 30–39 years, 2.4 per cent of women aged 40–
49 years and 0.5 per cent of women aged 50–60 years. The incidence of all acts of domestic
physical violence was also highest among younger women, with the exception of being
pushed or shoved (highest among women aged 25–29 years: 2.4 per cent) and being hit
(highest among women aged 20–24 years: 2.2 per cent). The association between age and
domestic physical violence was slightly different for men. The incidence of this form of
domestic violence was also highest (8.5 per cent) among younger men (15–19 years).
However, the lowest incidence of domestic physical violence was observed among men
aged 30–39 years (0.5 per cent). In contrast to the data discussed above for women, older
men aged 50–60 years were the most likely to report being hit (0.6 per cent) and being
choked or strangled (0.3 per cent).
Employment status. The incidence of domestic physical violence was highest among
women and men who were not working (9.1 and 3.8 per cent, respectively). The lowest
incidence was observed among women who were employed (4.7 per cent) but among men
who were self-employed (1.5 per cent).
Marital status. Women and men who had never been married at the time of the survey
were the most likely group to report having experienced at least one act of domestic
physical violence in the 12 months prior to the survey (9.0 and 5.6 per cent, respectively).
There were not substantial differences in the number of women and men who were
married or were divorced, separated or widowed and reported having experienced
domestic physical violence. However, women who were divorced, separated or widowed
were more likely than married or never-married women to experience acts of violence:
being pushed or shoved (2.2 per cent), being hit (1.5 per cent), being kicked, dragged or
beaten up (2.9 per cent) or being chocked or strangled (0.9 per cent). These results suggest
that these forms of violence were more likely to be perpetrated within intimate
partnerships. These patterns were observed for women only.
Geographical patterns. Both women and men living in rural areas were more likely to have
experienced domestic physical violence (6.4 and 2.6 per cent, respectively), than those in
urban areas (5.7 and 1.6 per cent, respectively). The highest incidence of domestic physical
violence among women was in the Eastern region (8.7 per cent, in comparison with the
lowest level of 0.8 per cent in the Upper East region). The highest incidence levels among
men were observed in the Central region (4.4 per cent), and the lowest in the Northern
region (0.7 per cent).
123
Education levels and asset ownership. The incidence of domestic physical violence was
highest among women with primary levels of education: 9.2 per cent of these women
experienced domestic physical violence, in comparison to, for instance, 2.9 per cent of
women with no education (who reported the lowest incidence of domestic physical
violence). Women in the middle asset quintile (8.2 per cent) reported higher levels of
domestic physical violence than women in the highest quintile (5.8 per cent) and women
in the lowest quintile (6.3 per cent). These results seem to suggest that women with
intermediate levels of education and wealth may be at a higher risk of experiencing
domestic physical violence. Men with secondary education were more likely to experience
domestic physical violence (3.6 per cent). The lowest levels of incidence were observed
among men with technical education (no men reported domestic physical violence), men
with no education (0.6 per cent) and men with higher education (0.8 per cent). The pattern
for wealth was clearer: men in the two lowest asset quintiles (3.2 and 3.5 per cent,
respectively) were more than three times more likely to experience domestic physical
violence than men in the fourth quintile (0.8 per cent) or in the top asset quintile (1.0 per
cent).
124
Table 31: Percentage of women who experienced domestic physical violence over the last 12 months
Slapped or
thrown
things at
Pushed
or
shoved
Hit Kicked,
dragged or
beaten up
Choked or
strangled
Burnt Attacked
with
weapon
Any
physical
violence
Age group (years)
15–19 10.4 1.5 1.2 4.9 1.3 1.1 1.1 15.2 20–24 7.2 1.5 2.2 2.1 0.0 0.3 0.0 9.4 25–29 5.5 2.4 1.5 2.6 1.1 0.3 1.0 6.9 30–39 4.7 1.2 1.4 2.4 0.6 0.0 1.0 7.1 40–49 2.0 0.8 0.8 0.3 0.2 0.0 0.2 2.4 50–60 0.5 0.0 0.0 0.0 0.0 0.0 0.0 0.5
Employment
Self-employed 3.5 1.2 1.1 1.8 0.5 0.0 0.4 5.1 Employed 3.2 1.1 1.4 1.4 0.4 0.0 0.7 4.7 Not working 7.3 1.5 1.3 2.0 0.4 0.7 1.0 9.1
Marital status
Never married 5.9 0.8 0.0 2.3 0.0 1.3 1.3 9.0 Married or living together 4.5 1.0 1.2 1.4 0.4 0.1 0.5 5.9 Divorced/separated/widowed 3.7 2.2 1.5 2.9 0.9 0.0 0.7 5.7
Residence
Urban 4.4 1.0 1.3 1.7 0.4 0.1 0.6 5.7 Rural 4.4 1.5 1.0 1.9 0.6 0.2 0.5 6.4
125
Slapped or
thrown
things at
Pushed
or
shoved
Hit Kicked,
dragged or
beaten up
Choked or
strangled
Burnt Attacked
with
weapon
Any
physical
violence
Region
Western 4.5 0.7 0.5 0.9 0.0 0.0 0.4 5.1 Central 5.0 1.4 1.0 2.2 0.5 0.5 0.8 6.9 Greater Accra 4.2 1.3 1.0 1.3 0.2 0.0 0.8 4.8 Volta 5.4 2.1 1.0 1.9 0.3 1.0 0.0 6.3 Eastern 6.3 1.7 2.5 2.8 1.1 0.0 1.2 8.7 Ashanti 4.4 0.8 1.2 2.1 0.2 0.0 0.4 6.7 Brong Ahafo 4.2 2.0 2.0 1.8 1.2 0.2 0.7 6.8 Northern 3.0 0.5 0.3 1.5 0.5 0.0 0.0 4.2 Upper East 0.0 0.8 0.8 0.8 0.8 0.0 0.0 0.8 Upper West 2.4 0.5 0.5 2.3 1.6 0.0 1.2 6.4
Education level
None 2.3 0.7 0.8 0.4 0.6 0.0 0.3 2.9 Primary 6.1 2.3 1.7 4.0 1.0 0.2 0.8 9.2 Middle/JSS/JHS 4.8 1.3 1.3 1.8 0.3 0.3 0.7 6.7 Secondary 5.1 0.9 1.5 1.7 0.5 0.0 0.0 6.6 Technical 4.5 3.5 0.6 4.0 0.0 0.0 3.5 5.1 Higher 3.4 0.2 0.2 0.3 0.0 0.0 0.2 3.8
Asset quintile
Lowest 4.0 2.1 1.3 2.0 1.3 0.4 0.3 6.3 Second 4.6 1.3 1.3 2.0 0.5 0.1 0.7 6.6 Middle 5.5 0.6 1.8 2.3 0.1 0.0 0.7 8.2 Fourth 3.5 0.6 0.8 0.9 0.4 0.2 0.2 3.7 Highest 4.8 1.7 0.7 2.0 0.0 0.0 1.1 5.8
Total 4.4 1.2 1.2 1.8 0.5 0.2 0.6 6.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
126
Table 32: Percentage of men who experienced domestic physical violence over the last 12 months
Slapped or
thrown
things at
Pushed
or
shoved
Hit Kicked,
dragged or
beaten up
Choked or
strangled
Burnt Attacked
with
weapon
Any
physical
violence
Age group (years)
15–19 6.4 0.7 0.5 0.7 0.0 0.8 0.0 8.5 20–24 1.9 0.0 1.0 0.3 0.0 0.0 0.3 3.4 25–29 1.2 0.0 0.2 0.4 0.2 0.0 0.4 2.0 30–39 0.3 0.0 0.0 0.0 0.0 0.0 0.3 0.5 40–49 0.4 0.3 0.0 0.0 0.0 0.0 0.0 0.6 50–60 0.7 0.6 0.6 0.3 0.3 0.0 0.3 0.9
Employment
Self-employed 0.9 0.3 0.2 0.2 0.2 0.0 0.5 1.5 Employed 1.6 0.0 0.6 0.0 0.0 0.0 0.0 2.0 Not working 2.4 0.3 0.2 0.6 0.0 0.4 0.0 3.8
Marital status
Never married 3.8 0.4 0.9 0.4 0.0 0.5 0.0 5.6 Married or living together 0.8 0.2 0.3 0.3 0.1 0.0 0.4 1.4 Divorced/separated/widowed 1.5 0.0 0.0 0.0 0.0 0.0 0.0 1.5
Residence
Urban 1.0 0.1 0.3 0.0 0.0 0.2 0.2 1.6 Rural 1.8 0.3 0.3 0.4 0.2 0.0 0.3 2.6
127
Slapped or
thrown
things at
Pushed
or
shoved
Hit Kicked,
dragged or
beaten up
Choked or
strangled
Burnt Attacked
with
weapon
Any
physical
violence
Region
Western 1.5 0.0 0.0 0.0 0.0 0.0 0.0 1.5 Central 1.1 0.7 0.6 0.0 0.6 1.2 1.4 4.4 Greater Accra 0.7 0.0 0.6 0.0 0.0 0.0 0.4 1.6 Volta 1.2 0.0 0.5 1.0 0.0 0.0 0.0 2.2 Eastern 3.6 0.0 0.2 0.0 0.0 0.0 0.0 3.8 Ashanti 1.4 0.0 0.2 0.5 0.0 0.0 0.0 2.1 Brong Ahafo 1.5 0.0 0.0 0.0 0.0 0.0 0.3 1.8 Northern 0.7 0.7 0.0 0.0 0.0 0.0 0.0 0.7 Upper East 1.4 1.1 1.1 1.1 1.1 0.0 0.0 1.8 Upper West 1.5 1.7 1.7 0.0 0.0 0.0 1.7 3.2
Education level
None 0.6 0.3 0.6 0.3 0.6 0.0 0.0 0.6 Primary 1.5 0.7 0.3 0.0 0.0 0.7 0.3 3.1 Middle/JSS/JHS 1.6 0.0 0.1 0.2 0.0 0.0 0.3 2.3 Secondary 2.5 0.3 0.8 0.6 0.0 0.0 0.4 3.6 Technical 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Higher 0.5 0.0 0.0 0.0 0.0 0.0 0.0 0.8
Asset quintile
Lowest 2.7 0.4 0.7 0.5 0.2 0.0 0.2 3.2 Second 1.5 0.0 0.6 0.5 0.2 0.3 0.5 3.5 Middle 1.3 0.4 0.0 0.1 0.0 0.0 0.0 1.5 Fourth 0.5 0.3 0.0 0.0 0.0 0.0 0.0 0.8 Highest 0.8 0.0 0.2 0.0 0.0 0.0 0.5 1.0
Total 1.4 0.2 0.3 0.2 0.1 0.1 0.2 2.1
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
128
4.2.3. Domestic sexual violence
Most types of sexual violence against women were predominantly perpetrated by
domestic relations, with the exception of sexual comments and sexual touches (3.1 and 5.7
per cent of these two acts, respectively, were committed by non-domestic relations). In
contrast, most acts of sexual violence against men – with the exception of sex without
consent and sex without protection – were mainly perpetrated by non-domestic relations.
This finding suggests that sexual violence against women was more prevalent within
domestic relations, whereas for men it was more widespread outside domestic relations
(Table 33). Most domestic sexual violence against women was perpetrated by non-family
members. This includes sexual comments (2.8 per cent perpetrated by non-family
members, and 0.3 per cent by family members), sexual touch (5.7 per cent perpetrated by
non-family members, and none by family members), physically forced to have sex (28.6
per cent perpetrated by non-family members, and 22.9 per cent by family members), and
having sex without protection (69.6 per cent perpetrated by non-family members, and 21.7
per cent perpetrated by family members). All acts of domestic sexual violence against men
were more likely to be committed by non-family members than family members (Table 33).
Table 33: Percentage of domestic sexual violence over the last 12 months
Sexual
comment
Sexual
touch
Physically
forced to
have sex
Otherwise
forced to
have sex
Sex
without
consent
Sex
without
protection
Sex
because
afraid
Women
Perpetrated by family
members
0.3 0.0 22.9 28.6 37.5 21.7 53.6
Perpetrated by non-
family members but
domestic
2.8
5.7
28.6
14.3
37.5
69.6
25.0
% Domestic violence 3.1 5.7 51.4 42.9 75.0 91.3 78.6
% Non-domestic
violence
96.9 94.3 48.6 57.1 25.0 8.7 21.4
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Men Perpetrated by family
members
1.4 1.4 0.0 0.0 0.0 0.0 6.7
Perpetrated by non-
family members but
domestic
2.2
4.3
12.0
18.8
60.0
75.0
60.0
% Domestic violence 3.6 5.7 12.0 18.8 60.0 75.0 66.7 % Non-domestic
violence
96.4 94.3 88.0 81.2 40.0 25.0 33.3
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Tables 34 and 35 illustrate, respectively, how domestic sexual violence was experienced
across different socio-economic groups. As with other forms of domestic violence, women
129
were more likely than men to have experienced domestic sexual violence (2.5 and 1.4 per
cent, respectively). The most common type of domestic sexual violence experienced by
women and men was being forced to have sex because of being afraid of the consequences
of refusing. This form of violence affected 0.8 per cent of all women in the GFLHS 2015 and
0.6 per cent of men. The second most common reason for both women and men was being
forced to have sex without protection (0.7 per cent for women and 0.3 per cent for men).
Age patterns. In contrast to other forms of domestic violence, the incidence of domestic
sexual violence was highest among women aged 20–24 years: 4.9 per cent of women in this
age group reported having experienced domestic sexual violence, in comparison to 4.6 per
cent of women aged 25–29 years, 4.3 per cent of women aged 15–19 years, 2.2 per cent of
women aged 30–39 years and 0.5 per cent of women aged 40–49 years. However, the
incidence of particular types of domestic sexual violence was highest among young
women aged 15–19 years: unwanted sexual touch (0.7 per cent), being physically forced to
have sex (1.5 per cent), and being forced to have sex without protection (1.6 per cent). As
with other forms of domestic violence, the incidence of domestic sexual violence was
lowest among the older women (aged 50–60 years). Men aged 30–39 were at the highest
risk of experiencing domestic sexual violence (2.7 per cent) across almost all acts of sexual
violence reported. The main exception was being the target of an unwanted sexual
comment, which was most prevalent (0.7 per cent) among the younger age group (15–19
years). No men aged 50–60 years reported having experienced an act of sexual violence in
the 12 months prior to the survey.
Employment status. Women who were either employed or not working at the time of the
survey were most likely to report domestic sexual violence (3.0 per cent in both cases).
Employed men were most at risk of domestic sexual violence (also 3.0 per cent). The results
in Tables 34 and 35 show, however, great variation in the association between employment
status and the incidence of domestic violence, with no clear pattern emerging from the
data.
Marital status. The incidence of domestic sexual violence was highest among married
women (2.9 per cent) and divorced, separated or widowed men (2.1 per cent). No never-
married women reported having experienced domestic sexual violence. The incidence of
domestic sexual violence was lowest among never-married men (0.4 per cent). The results
show, however, a variation in how women and men experienced different acts of domestic
sexual violence. Notably, women who were divorced, separated or widowed experienced
the highest incidence of unwanted sexual comments (0.6 per cent) and acts of forced sex
(0.2 per cent). Married men experienced the highest incidence of unwanted sexual touches
(0.2 per cent) and acts of physically forced sex (0.3 per cent), while never-married men
reported the highest incidence of unwanted sexual comments (0.4 per cent).
Geographical patterns. Similar to domestic physical violence, women and men living in
urban areas reported a higher incidence of domestic sexual violence (2.8 and 1.6 per cent,
respectively) than those in rural areas. The highest level of incidence of domestic sexual
130
violence among women and men was observed in the Central region (5.8 and 2.5 per cent,
respectively). The lowest incidence of domestic sexual violence among women was in the
Northern and Upper East regions (both 0.6 per cent), and no men in the Brong Ahafo,
Upper East or Upper West regions reported domestic sexual violence.
Education levels and asset ownership. The incidence of domestic sexual violence was
highest among women with secondary education: 4.9 per cent of these women experienced
at least one act of sexual violence in the 12 months prior to the survey, in comparison to,
for instance, 0.9 per cent of women with no education. Men with technical or higher
education were more likely to have experienced domestic sexual violence (2.2 and 2.1 per
cent, respectively, in comparison with men with no education). Both men and women in
the two highest asset quintiles were more at risk of experiencing domestic sexual violence
(3.9 and 2.5 per cent for women, and 1.7 and 2.7 per cent for men) than men and women in
the other asset quintiles.
131
Table 34: Percentage of women who experienced domestic sexual violence over the last 12 months
Sexual
comment
Sexual
touch
Physically
forced to
have sex
Otherwise
forced to
have sex
Sex
without
consent
Sex
without
protection
Had sex
because
afraid
Any
sexual
violence
Age group (years)
15–19 0.0 0.7 1.5 0.0 0.4 1.6 0.5 4.3 20–24 0.3 0.3 1.1 0.0 1.2 1.1 1.7 4.9 25–29 0.4 0.6 0.9 0.5 0.5 1.0 1.5 4.6 30–39 0.0 0.0 0.6 0.2 0.7 0.8 0.7 2.2 40–49 0.1 0.0 0.1 0.0 0.1 0.0 0.4 0.5 50–60 0.3 0.3 0.0 0.0 0.0 0.0 0.0 0.3
Employment
Self-employed 0.2 0.2 0.7 0.1 0.6 0.6 0.6 2.3 Employed 0.6 0.3 0.5 0.0 0.0 0.9 0.8 3.0 Not working 0.0 0.3 0.4 0.2 0.5 0.8 1.4 3.0
Marital status
Never married 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Married or living together 0.1 0.3 0.8 0.1 0.7 0.7 1.0 2.9 Divorced/separated/widowed 0.6 0.3 0.1 0.2 0.0 0.7 0.3 1.9
Residence
Urban 0.2 0.2 0.5 0.0 0.3 0.9 1.2 2.8 Rural 0.2 0.3 0.7 0.3 0.7 0.4 0.4 2.2
132
Sexual
comment
Sexual
touch
Physically
forced to
have sex
Otherwise
forced to
have sex
Sex
without
consent
Sex
without
protection
Had sex
because
afraid
Any
sexual
violence
Region
Western 0.1 0.4 0.2 0.0 0.3 0.3 0.3 1.1 Central 0.0 0.5 1.4 0.5 0.9 1.4 2.0 5.8 Greater Accra 0.2 0.0 0.2 0.0 0.8 1.2 1.6 3.2 Volta 0.0 0.6 1.3 0.6 0.0 0.3 1.0 3.4 Eastern 0.8 0.8 0.6 0.0 0.6 0.6 0.6 2.2 Ashanti 0.3 0.3 0.8 0.3 0.7 0.1 0.6 2.8 Brong Ahafo 0.0 0.0 0.3 0.0 0.0 1.0 0.0 1.0 Northern 0.0 0.0 0.6 0.0 0.3 0.0 0.0 0.6 Upper East 0.0 0.0 0.0 0.0 0.0 0.6 0.0 0.6 Upper West 0.0 0.0 0.0 0.0 0.0 1.1 0.0 1.1
Education level
None 0.1 0.1 0.5 0.2 0.1 0.1 0.0 0.9 Primary 0.1 0.0 0.5 0.0 0.7 0.1 1.0 2.2 Middle/JSS/JHS 0.3 0.4 0.8 0.2 0.8 0.9 0.8 2.9 Secondary 0.5 0.3 0.9 0.0 0.3 1.5 1.8 4.9 Technical 0.0 0.0 0.0 0.0 0.0 0.0 1.9 1.9 Higher 0.0 0.8 0.0 0.0 0.0 1.2 0.5 2.5
Asset quintile
Lowest 0.0 0.2 0.5 0.2 0.0 0.3 0.3 1.6 Second 0.3 0.5 0.7 0.0 0.6 0.6 0.6 2.3 Middle 0.1 0.0 0.9 0.3 0.4 0.7 0.4 2.2 Fourth 0.2 0.0 0.7 0.2 1.0 0.9 1.7 3.9 Highest 0.3 0.5 0.2 0.0 0.3 0.8 1.0 2.5
Total 0.2 0.3 0.6 0.1 0.5 0.7 0.8 2.5
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
133
Table 35: Percentage of men who experienced domestic sexual violence over the last 12 months
Sexual
comment
Sexual
touch
Physically
forced to
have sex
Otherwise
forced to
have sex
Sex
without
consent
Sex
without
protection
Had sex
because
afraid
Any
sexual
violence
Age group (years)
15–19 0.7 0.0 0.0 0.0 0.0 0.0 0.0 0.7 20–24 0.0 0.0 1.0 0.0 0.3 0.5 1.2 2.3 25–29 0.3 0.2 0.0 0.4 0.4 0.0 0.5 1.4 30–39 0.3 0.1 0.2 0.5 0.3 0.7 1.1 2.7 40–49 0.0 0.2 0.0 0.0 0.0 0.2 0.0 0.3 50–60 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Employment
Self-employed 0.0 0.1 0.3 0.1 0.1 0.0 0.2 0.8 Employed 0.2 0.1 0.2 0.5 0.4 0.5 1.7 3.0 Not working 0.6 0.0 0.0 0.0 0.0 0.6 0.0 1.0
Marital status
Never married 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.4 Married or living together 0.1 0.2 0.3 0.2 0.1 0.2 0.6 1.5 Divorced/separated/widowed 0.3 0.0 0.0 0.4 0.7 0.7 0.9 2.1
Residence
Urban 0.1 0.2 0.1 0.4 0.2 0.2 0.6 1.6 Rural 0.2 0.0 0.3 0.0 0.2 0.3 0.5 1.2
134
Sexual
comment
Sexual
touch
Physically
forced to
have sex
Otherwise
forced to
have sex
Sex
without
consent
Sex
without
protection
Had sex
because
afraid
Any
sexual
violence
Region
Western 0.0 0.2 0.0 0.5 0.0 0.7 0.7 2.1 Central 0.8 0.4 0.0 0.0 0.0 0.7 1.2 2.5 Greater Accra 0.0 0.1 0.3 0.8 0.4 0.2 1.0 2.3 Volta 0.1 0.0 1.5 0.0 0.5 0.0 0.5 1.6 Eastern 0.0 0.0 0.0 0.0 0.7 0.0 1.8 1.8 Ashanti 0.6 0.0 0.0 0.0 0.0 0.2 0.0 0.8 Brong Ahafo 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Northern 0.0 0.3 0.0 0.0 0.0 0.8 0.0 1.1 Upper East 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Upper West 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Education level
None 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Primary 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.2 Middle/JSS/JHS 0.2 0.1 0.0 0.3 0.0 0.3 0.9 1.8 Secondary 0.0 0.2 0.7 0.0 0.2 0.3 0.8 1.7 Technical 0.0 0.0 0.0 2.2 2.2 0.0 0.0 2.2 Higher 0.8 0.2 0.3 0.0 0.4 0.6 0.4 2.1
Asset quintile
Lowest 0.3 0.0 0.2 0.0 0.2 0.6 0.2 1.2 Second 0.0 0.1 0.0 0.0 0.0 0.0 0.8 0.9 Middle 0.0 0.0 0.0 0.0 0.0 0.2 0.7 0.9 Fourth 0.5 0.0 0.0 0.7 0.4 0.3 0.4 1.7 Highest 0.1 0.4 0.9 0.3 0.4 0.4 0.7 2.7
Total 0.2 0.1 0.2 0.2 0.2 0.3 0.6 1.4
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
135
4.2.4. Domestic psychological violence
Table 36 shows that most psychological violence against women was perpetrated within
domestic relations, with the exception of insults (37.1 per cent of this violence was
perpetrated by domestic relations, and 62.9 per cent by non-domestic relations). Most
psychological violence against men was perpetrated outside domestic relations, with the
exception of acts of being ignored or threatened with abandonment (59.6 per cent of this
violence was perpetrated by domestic relations) (Table 36). All acts of psychological
violence against women and men were more likely to be perpetrated by non-family
members than by family members.
Table 36: Percentage of domestic psychological violence over the last 12 months
Insulted,
humiliated,
belittled
Ignored,
threatened
to be
abandoned
Scared,
intimidated
Attacked
with
weapon
Threaten to
be hurt
Women
Perpetrated by family members 13.4 33.7 12.5 26.1 14.3
Perpetrated by non-family
members but domestic
24.2
46.9
37.5
52.2
50.0
% Domestic violence 37.1 80.6 50.0 78.3 64.3
% Non-domestic violence 62.9 19.4 50.0 21.7 35.7
Total 100.0 100.0 100.0 100.0 100.0
Men Perpetrated by family members 7.6 17.3 2.8 0.0 5.6
Perpetrated by non-family
members but domestic
20.7
42.3
8.3
15.0
16.7
% Domestic violence 27.6 59.6 11.1 15.0 22.2
% Non-domestic violence 72.4 40.4 88.9 85.0 77.8
Total 100.0 100.0 100.0 100.0 100.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Tables 37 and 38 disaggregate the information above further and show how the incidence
of domestic psychological violence was distributed across different socio-economic
characteristics and geographical locations. As with other types of domestic violence,
women were more likely to experience domestic psychological violence (6 per cent) than
men (2.1 per cent). The most common act of domestic psychological violence was being
insulted, humiliated or belittled (7.5 per cent for women and 6.9 per cent for men). The
second most common act was being ignored, or threatened with abandonment (2.4 per cent
for women and 1.5 per cent for men).
Age patterns. The incidence of domestic psychological violence was highest among women
and men in the youngest age group (15–19 years old): 15.2 per cent of women and 8.5 per
cent of men in this age group experienced at least one act of domestic psychological
136
violence in the 12 months prior to the survey. Levels of incidence of domestic psychological
violence reduced as the age of women increased. Among the older age group (50–60 years),
only 0.5 per cent of women reported having experienced domestic psychological violence.
However, the incidence of certain forms of psychological violence, such as being
threatened with a weapon, was highest among women aged 25–29 years (0.9 per cent),
whereas being threatened to be hurt was highest among women aged 30–39 years. The
lowest incidence of domestic psychological violence among men was among those aged
30–39 years (0.5 per cent).
Employment status. Women and men who were not working at the time of the survey
were more likely to experience at least one act of domestic psychological violence (9.1 per
cent for women and 3.8 per cent for men). The lowest incidence of domestic psychological
violence was observed among employed women (4.7 per cent) and self-employed men (1.5
per cent).
Marital status. The incidence of domestic psychological violence was highest among
women and men who were never married (9.0 and 5.6 per cent, respectively). However,
divorced, separated or widowed women were the most likely group to report the incidence
of particular acts of domestic psychological violence, such as being threatened with the use
of a weapon (0.8 per cent) or being threatened to be hurt (0.7 per cent). Likewise, divorced,
separated or widowed men were the most likely to report being insulted, humiliated or
belittled (9.6 per cent) or being threatened to be hurt (0.7 per cent), while married men were
the most likely to report being scared or intimidated on purpose (0.3 per cent) or being
threatened with a weapon (0.3 per cent).
Geographical patterns. Women and men living in rural areas were more likely to have
reported being a victim of domestic psychological violence (6.4 and 2.6 per cent,
respectively) than those in urban areas (5.7 and 1.6 per cent, respectively). The incidence
of domestic psychological violence among women was highest in the Eastern region (8.7
per cent) and lowest in the Upper East region (0.8 per cent). Among men, it was highest
among those living in the central region (4.4 per cent) and lowest among those living in the
Northern region (0.7 per cent).
Education levels and asset ownership. The incidence of domestic psychological violence
was highest among women with primary levels of education (9.2 per cent), and lowest
among women with no education (2.9 per cent) or women with the highest levels of
education (3.8 per cent). Men with secondary education were those most likely to report
experiences of domestic psychological violence (3.6 per cent). As with women, men with
no education (0.6 per cent) or men with the highest levels of education (technical: none,
and higher: 0.8 per cent) were the least likely to report having experienced at least one act
of domestic psychological violence in the 12 months prior to the survey. The incidence of
domestic psychological violence was highest among women in the middle asset quintile
(8.2 per cent) and men in the second asset quintile (3.5 per cent). Women and men in the
fourth and fifth asset quintiles were the least likely to report experiences of domestic
137
psychological violence (3.7 and 5.8 per cent, respectively, for women, and 0.8 and 1.0 per
cent for men). These results suggest that women and men at intermediate levels of
education and wealth were those most at risk of experiencing domestic psychological
violence.
138
Table 37: Percentage of women who experienced domestic psychological violence over the last 12 months
Insulted,
humiliated,
belittled
Ignored,
threatened
to be
abandoned
Scared,
intimidated
Attacked
with
weapon
Threaten
to be hurt
Any
psychological
violence
Age group (years)
15–19 20.0 6.8 1.4 1.0 0.0 15.2 20–24 10.5 4.2 0.3 0.0 0.2 9.4 25–29 8.5 2.3 0.3 0.9 0.3 6.9 30–39 7.0 2.1 0.5 0.7 0.6 7.1 40–49 3.5 1.9 0.2 0.6 0.1 2.4 50–60 4.3 0.7 0.6 0.6 0.0 0.5
Employment
Self-employed 6.5 2.0 0.4 0.5 0.3 5.1 Employed 5.8 2.4 0.4 1.2 0.0 4.7 Not working 11.2 3.7 0.8 0.9 0.4 9.1
Marital status
Never married 15.8 3.5 1.2 0.8 0.0 9.0 Married or living together 6.9 2.3 0.4 0.6 0.2 5.9 Divorced/separated/widowed 7.8 2.8 0.8 0.8 0.7 5.7
Residence
Urban 6.7 2.7 0.4 0.4 0.2 5.7 Rural 8.5 2.2 0.6 0.9 0.3 6.4
139
Insulted,
humiliated,
belittled
Ignored,
threatened
to be
abandoned
Scared,
intimidated
Attacked
with
weapon
Threaten
to be hurt
Any
psychological
violence
Region
Western 6.8 0.9 0.0 0.4 0.3 5.1 Central 8.8 4.8 0.5 0.0 0.5 6.9 Greater Accra 6.7 2.9 0.2 0.8 0.2 4.8 Volta 8.8 1.2 2.0 0.4 0.0 6.3 Eastern 11.3 4.0 2.2 2.2 0.0 8.7 Ashanti 7.3 2.3 0.0 0.5 0.3 6.7 Brong Ahafo 7.4 3.0 0.0 0.7 0.7 6.8 Northern 5.8 0.7 0.0 0.0 0.3 4.2 Upper East 1.0 0.0 0.0 0.0 0.0 0.8 Upper West 7.2 0.5 0.0 0.0 0.0 6.4
Education level
None 5.1 1.6 0.6 0.3 0.3 2.9 Primary 9.4 3.5 0.9 1.4 0.7 9.2 Middle/JSS/JHS 8.2 2.4 0.3 0.5 0.1 6.7 Secondary 7.9 3.5 0.6 0.2 0.3 6.6 Technical 8.9 2.0 0.0 3.5 0.0 5.1 Higher 5.9 1.5 0.2 0.2 0.2 3.8
Asset quintile
Lowest 10.1 2.3 1.1 1.2 0.6 6.3 Second 6.1 2.9 0.3 0.6 0.3 6.6 Middle 9.1 2.8 0.2 0.3 0.3 8.2 Fourth 7.5 1.6 0.3 0.1 0.2 3.7 Highest 5.3 2.5 0.6 1.1 0.0 5.8
Total 7.5 2.4 0.5 0.6 0.3 6.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
140
Table 38: Percentage of men who experienced domestic psychological violence over the last 12 months
Insulted,
humiliated,
belittled
Ignored,
threatened
to be
abandoned
Scared,
intimidated
Attacked
with
weapon
Threaten
to be hurt
Any
psychological
violence
Age group (years)
15–19 11.5 3.2 0.7 0.0 0.0 8.5 20–24 10.2 3.3 0.0 0.3 0.4 3.4 25–29 7.8 0.8 0.2 0.0 0.4 2.0 30–39 4.2 1.0 0.0 0.3 0.5 0.5 40–49 6.7 1.1 0.3 0.0 0.0 0.6 50–60 3.8 0.9 0.3 0.3 0.0 0.9
Employment
Self-employed 6.1 1.6 0.1 0.3 0.2 1.5 Employed 6.9 1.0 0.4 0.0 0.6 2.0 Not working 8.8 2.2 0.3 0.0 0.0 3.8
Marital status
Never married 8.3 3.1 0.0 0.0 0.0 5.6 Married or living together 5.8 1.1 0.3 0.3 0.2 1.4 Divorced/separated/widowed 9.6 1.5 0.0 0.0 0.7 1.5
Residence
Urban 6.4 1.5 0.0 0.2 0.4 1.6 Rural 7.4 1.6 0.4 0.2 0.1 2.6
141
Insulted,
humiliated,
belittled
Ignored,
threatened
to be
abandoned
Scared,
intimidated
Attacked
with
weapon
Threaten
to be hurt
Any
psychological
violence
Region
Western 4.2 0.6 0.0 0.0 0.0 1.5 Central 8.2 5.2 0.7 0.7 0.0 4.4 Greater Accra 6.4 0.9 0.0 0.4 0.7 1.6 Volta 9.5 1.5 0.0 0.0 0.0 2.2 Eastern 14.2 2.9 0.0 0.0 0.6 3.8 Ashanti 8.6 1.5 0.0 0.0 0.4 2.1 Brong Ahafo 2.5 1.1 0.4 0.4 0.0 1.8 Northern 5.2 1.1 0.7 0.0 0.0 0.7 Upper East 1.1 0.6 1.1 0.0 0.0 1.8 Upper West 0.0 0.0 0.0 0.0 0.0 3.2
Education level
None 3.9 0.8 0.0 0.0 0.0 0.6 Primary 8.1 2.9 0.6 0.0 0.4 3.1 Middle/JSS/JHS 7.2 1.3 0.2 0.2 0.0 2.3 Secondary 8.2 2.7 0.0 0.4 0.0 3.6 Technical 4.2 1.5 0.0 0.0 0.0 0.0 Higher 6.5 0.0 0.3 0.0 1.3 0.8
Asset quintile
Lowest 6.1 1.7 0.0 0.0 0.0 3.2 Second 7.2 2.5 0.6 0.3 0.0 3.5 Middle 7.9 1.8 0.3 0.0 0.7 1.5 Fourth 7.0 1.1 0.0 0.0 0.0 0.8 Highest 6.2 0.0 0.0 0.5 0.7 1.0
Total 6.9 1.5 0.2 0.2 0.2 2.1
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
142
4.2.5. Domestic economic violence
For most acts of economic violence against women, the majority were perpetrated by
domestic relations, with the exception of cash being taken and property damaged: 34.4 and
46.4 per cent of these acts, respectively, were committed by domestic relations (Table 39).
A large part of economic violence against men was perpetrated outside domestic relations,
with three exceptions: 83.2 per cent of violence in the form of being denied pocket money,
92.3 per cent of violence in the form of having belongings controlled, and 52.2 per cent of
violence in the form of being prohibited from working were perpetrated by domestic
relations (Table 39). All forms of domestic economic violence against men were more likely
to be committed by non-family members than by family members. Most forms of domestic
economic violence against women were also committed by non-family members, with the
exception of having their belongings controlled (60.9 per cent committed by family
members, and 34.8 per cent by non-family members) and being prohibited from working
(48.6 per cent committed by family members, and 28.6 per cent by non-family members).
Table 39: Percentage of domestic economic violence over the last 12 months
Denied
money
Cash
taken out
Belongings
controlled
Property
damaged
Prohibited
from
working
Forced to
work
Denied
food
Women
Perpetrated by family
members
44.1 15.6 60.9 3.6 48.6 30.0 36.4
Perpetrated by non-
family members but
domestic
52.3
18.8
34.8
42.9
28.6
33.3
59.1
% Domestic violence 96.0 34.4 91.3 46.4 74.3 63.3 95.4
% Non-domestic
violence
4.0 65.6 8.7 53.6 25.7 36.7 4.6
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Men
Perpetrated by family
members
28.7 15.1 38.5 5.7 0.0 7.7 23.1
Perpetrated by non-
family members but
domestic
54.5
18.2
53.8
15.1
52.2
25.6
50.0
% Domestic violence 83.2 33.3 92.3 18.9 52.2 33.3 73.1
% Non-domestic
violence
16.8 66.7 7.7 81.1 47.8 66.7 26.9
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Tables 40 and 41 illustrate how domestic economic violence was distributed across the
socio-economic characteristics and geographical location of women and men in Ghana.
Domestic economic violence was experienced by 12.8 per cent of women and 7.3 per cent
of men interviewed in the GFLHS 2015. The most common act of domestic economic
143
violence experienced by women and men was being denied pocket money – known in
Ghana as chop money. This act of domestic economic violence was experienced by 11.2 per
cent of women and 4.9 per cent of men. Less than 1 per cent of the sample reported having
experienced any other act of domestic economic violence.
Age patterns. The incidence of domestic economic violence overall was highest among
women aged 25–29 years (19.7 per cent), in comparison to 17.2 per cent among women
aged 15–19 years, 15.9 per cent among women aged 20–24 years, 13 per cent among women
aged 30–39 years, 8.1 per cent among women aged 40–49 years and 5.1 per cent among
women in the oldest age group (50–60 years old). However, women in the youngest age
group (15–19 years) were at a greater risk of particular acts of domestic violence. Notably,
being denied food was highest among this group (5.5 per cent) than among any other age
group (Table 40). These women also reported the highest incidence of being forced to work
(1.5 per cent) and having their belongings controlled (1.6 per cent). The incidence of
domestic economic violence among men was highest among the youngest age group (15–
19 years), at 19 per cent, in comparison to 3.4 per cent among the oldest age group (50–60
years).
Employment status. Women and men who were not working at the time of the survey
were more likely to report having experienced at least one act of domestic economic
violence in the 12 months prior to the survey (16.6 per cent of women and 15.3 per cent of
men). However, the incidence of certain types of domestic violence, such as being
prohibited from working, was highest among self-employed women (1.2 per cent),
whereas employed men were the most likely group to report having their cash taken (0.6
per cent) or being denied food (2.1 per cent). Employed women (8.7 per cent) and self-
employed men (3.9 per cent) were the least likely to report domestic economic violence.
Marital status. The association between marital status and the incidence of domestic
economic violence differs between women and men. Married women (13.4 per cent) were
more likely to experience domestic economic violence than never-married women (10.1
per cent) and women who were divorced, separated or widowed (11.6 per cent). The
incidence of domestic economic violence was highest among never-married men (13.4 per
cent), in comparison to 9.9 per cent among men who were divorced, separated or widowed
and 4.9 per cent among men who were married.
Geographical patterns. The incidence of domestic economic violence was higher among
women living in rural areas (15.9 per cent) than among those living in urban areas (10.4
per cent). The incidence of domestic economic violence was also higher among men living
in rural areas (7.4 per cent), but the difference in relation to men living in urban areas (7.3
per cent) was small. Women living in the Western region (22.8 per cent) and in the Upper
East region (6.7 per cent) were, respectively, the most likely and least likely to report
incidence of domestic economic violence. The incidence of domestic economic violence
was highest among men in the Central region (16.7 per cent) and lowest among men in the
Upper East region (2 per cent).
144
Education levels and asset ownership. Women with primary education were the most
likely to report experiences of domestic economic violence (16.7 per cent). The lowest
incidence of domestic economic violence was observed among women with either higher
education (8.1 per cent) or women with no education (9.6 per cent). Patterns for men were
similar: the incidence of domestic economic violence was lowest among men with no
education (1.4 per cent) or the highest level of education (5.6 per cent), and highest among
men with secondary education (9.6 per cent). Women and men in the second asset quintile
were the most likely to report having experienced domestic economic violence (15.3 and
9.1 per cent, respectively), while women and men in the top asset quintile had the lowest
incidence of domestic economic violence (6.9 and 5.8 per cent, respectively). These results
suggest that individuals with intermediate levels of education and wealth may be at higher
risk of economic violence than better-off and more educated individuals.
145
Table 40: Percentage of women who experienced domestic economic violence over the last 12 months
Denied
money
Cash
taken
out
Belongings
controlled
Property
damaged
Prohibited
from
working
Forced
to work
Denied
food
Any
economic
violence
Age group (years)
15–19 12.9 0.6 1.6 0.2 0.0 1.5 5.5 17.2 20–24 13.2 0.7 1.0 0.5 1.1 1.2 1.9 15.9 25–29 16.7 0.9 1.0 1.1 1.8 1.3 1.2 19.7 30–39 12.1 0.1 0.8 0.5 1.0 0.1 1.2 13.0 40–49 7.8 0.1 0.3 0.0 0.6 0.5 0.2 8.1 50–60 4.2 0.0 0.1 0.1 0.5 0.0 0.3 5.1
Employment
Self-employed 10.9 0.4 0.4 0.4 1.2 0.5 1.1 12.1 Employed 7.6 0.0 1.0 0.4 0.4 0.2 0.3 8.7 Not working 13.5 0.5 1.4 0.7 0.5 1.0 2.2 16.6
Marital status
Never married 6.1 0.0 0.9 0.0 0.0 1.9 3.7 10.1 Married or living together 12.0 0.5 0.7 0.4 1.1 0.5 0.9 13.4 Divorced/separated/widowed 9.7 0.0 0.7 0.8 0.7 0.6 1.9 11.6
Residence
Urban 8.8 0.3 0.9 0.4 0.6 0.4 0.7 10.4 Rural 13.8 0.5 0.5 0.5 1.4 0.9 1.9 15.6
146
Denied
money
Cash
taken
out
Belongings
controlled
Property
damaged
Prohibited
from
working
Forced
to work
Denied
food
Any
economic
violence
Region
Western 22.4 0.3 0.4 0.7 0.0 0.0 1.2 22.8 Central 10.8 0.9 0.9 1.1 0.4 0.5 2.4 12.8 Greater Accra 6.8 0.0 1.0 0.0 0.1 0.4 0.2 7.7 Volta 5.7 0.5 0.0 0.0 0.2 0.4 0.7 6.9 Eastern 15.7 0.9 1.1 0.8 0.9 0.8 1.3 18.3 Ashanti 12.2 0.4 0.4 0.5 1.3 0.0 1.4 14.2 Brong Ahafo 11.7 0.0 0.5 0.8 3.2 1.6 1.9 14.6 Northern 6.9 0.3 0.9 0.0 1.8 2.1 2.5 8.7 Upper East 6.1 0.0 1.1 0.0 0.0 0.0 0.6 6.7 Upper West 12.5 0.0 2.4 0.0 2.4 1.5 0.0 13.9
Education level
None 8.6 0.2 0.7 0.1 0.9 0.5 1.0 9.6 Primary 14.7 0.7 0.9 0.1 0.9 0.3 0.7 16.7 Middle/JSS/JHS 12.3 0.5 0.6 0.8 1.2 0.8 2.0 14.2 Secondary 10.2 0.0 0.4 0.9 1.0 0.6 0.9 11.9 Technical 9.8 0.0 0.0 0.0 1.0 0.0 0.8 10.8 Higher 6.7 0.0 1.4 0.0 0.0 0.5 0.0 8.1
Asset quintile
Lowest 11.1 0.3 0.5 0.2 1.4 1.6 2.2 13.4 Second 13.7 0.3 0.8 0.7 1.3 0.7 1.2 15.3 Middle 14.5 0.1 0.8 0.5 0.3 0.0 2.0 15.2 Fourth 9.2 0.7 1.0 0.5 1.0 0.5 0.5 11.6 Highest 6.0 0.2 0.5 0.2 0.5 0.0 0.4 6.9
Total 11.2 0.4 0.7 0.4 1.0 0.6 1.2 12.8
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
147
Table 41: Percentage of men who experienced domestic economic violence over the last 12 months
Denied
money
Cash
taken
out
Belongings
controlled
Property
damaged
Prohibited
from
working
Forced
to work
Denied
food
Any
economic
violence
Age group (years)
15–19 14.3 1.2 2.9 0.9 0.5 2.4 1.6 19.0 20–24 8.9 1.3 0.1 1.5 1.5 1.8 3.0 13.6 25–29 4.8 0.6 0.0 0.4 0.8 1.2 0.9 7.3 30–39 1.8 0.0 0.4 0.1 0.2 0.0 0.0 2.5 40–49 2.3 0.2 0.6 0.2 0.9 0.3 0.8 4.9 50–60 2.5 0.0 0.8 0.8 0.3 0.0 0.6 3.4
Employment
Self-employed 2.4 0.4 0.1 0.1 0.6 0.6 0.3 3.9 Employed 4.2 0.6 0.9 0.6 0.8 1.1 2.1 7.5 Not working 12.0 0.5 1.6 1.7 0.8 1.0 1.4 15.3
Marital status
Never married 10.4 0.7 1.5 0.5 0.4 1.4 2.1 13.4 Married or living together 3.0 0.2 0.5 0.1 0.5 0.7 0.5 4.9 Divorced/separated/widowed 5.9 1.1 0.3 1.6 1.6 0.8 1.6 9.9
Residence
Urban 4.9 0.2 1.0 0.7 0.5 0.8 1.0 7.3 Rural 4.9 0.8 0.2 0.5 0.9 0.8 1.1 7.4
148
Denied
money
Cash
taken
out
Belongings
controlled
Property
damaged
Prohibited
from
working
Forced
to work
Denied
food
Any
economic
violence
Region
Western 7.7 0.2 0.3 0.2 0.0 0.0 0.5 8.7 Central 8.6 2.3 4.5 0.0 1.3 1.3 5.6 16.7 Greater Accra 5.3 0.0 1.1 0.0 0.9 1.5 0.6 6.3 Volta 2.0 0.0 0.0 0.0 0.0 0.0 0.8 2.3 Eastern 2.3 2.2 0.8 0.6 2.8 1.0 2.1 9.4 Ashanti 6.9 0.2 0.0 2.4 0.0 0.7 0.7 9.1 Brong Ahafo 4.4 0.5 0.0 0.3 1.3 0.0 0.0 6.5 Northern 2.2 0.0 0.0 0.3 0.0 1.8 0.3 4.6 Upper East 2.0 0.0 0.0 0.0 0.0 0.0 0.0 2.0 Upper West 2.2 0.0 0.0 0.0 2.2 2.2 0.0 2.2
Education level
None 0.6 0.0 0.0 0.2 0.0 0.6 0.3 1.4 Primary 4.7 0.7 1.0 0.0 1.7 0.6 1.7 8.6 Middle/JSS/JHS 5.8 0.7 0.9 0.4 1.0 0.9 1.0 8.7 Secondary 7.7 0.6 0.1 1.2 0.3 1.2 1.3 9.6 Technical 5.0 0.0 0.0 0.0 0.0 2.6 1.5 6.5 Higher 3.3 0.2 1.1 1.0 0.4 0.0 0.6 5.6
Asset quintile
Lowest 6.2 0.2 0.5 0.5 0.8 0.9 0.9 7.2 Second 5.0 0.7 0.5 0.7 1.4 1.5 1.7 9.1 Middle 5.3 0.7 0.3 0.7 0.0 0.6 1.0 7.4 Fourth 5.2 0.2 1.0 0.4 0.4 0.6 0.6 6.5 Highest 2.6 0.6 0.9 0.6 0.6 0.0 0.6 5.8
Total 4.9 0.5 0.6 0.6 0.7 0.8 1.0 7.3
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
149
4.2.6. Summary
The information discussed in the sections above suggests important patterns in how
domestic violence was experienced among men and women, and different socio-economic
groups and geographical locations in Ghana. Some key findings include:
Younger women and men were overall at higher risk than older age groups of
experiencing any form of domestic violence.
Domestic physical or psychological violence appears to have been more prevalent
among non-working, never-married men and women with intermediate levels of
education and wealth and living in rural areas.
Domestic economic violence exhibits a similar pattern as above but was more
prevalent among married women.
Domestic social violence was more prevalent among non-working and urban
women and men with some level of education and wealth.
Domestic sexual violence differs from most other forms of domestic violence: it
appears to be most prominent among employed, better-off, less economically
dependent individuals living in urban areas.
Table 42 summarises these findings, highlighting the categories for which each type of
violence was most prominent. These findings are explored in detail in the remainder of
Section 4.
Table 42: Socio-economic groups with highest incidence of domestic violence
Social
violence
Physical
violence
Sexual
violence
Psychological
violence
Economic
violence
Age (years)
Women 15–19 15–19 20–24 15–19 25–29
Men 15–19 15–19 30–39 15–19 15–19
Employment status
Women Not working Not working Employed;
not working
Not working Not working
Men Not working Not working Employed Not working Not working
Marital status
Women Not married Not married Married Not married Married
Men Divorced Not married Divorced Not married Not married
Residence
Women Urban Rural Urban Rural Rural
Men Urban Rural Urban Rural Rural
Education
Women Secondary Primary Secondary Primary Primary
Men Higher Secondary Technical;
higher
Secondary Secondary
Asset quintile
Women Third Third Fourth Third Second
Men Highest Second Highest Second Second
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
150
Qualitative data across all regions supported the findings discussed above: both men and
women experienced domestic violence, albeit in different forms and with significant
variations across individual, household, community and regional characteristics. The data
from both the focus group discussions and the in-depth interviews suggested that women
experienced two predominant forms of domestic violence: economic violence (in the form
of men’s refusal to provide chop money) and physical violence (in the form of wife-beating).
The qualitative data showed, in addition and in line with the quantitative results above,
that men experienced domestic violence, by women and sometimes by other men, largely
in more public spaces – streets or marketplaces. The forms of domestic violence that men
experienced in these spaces were predominantly verbal abuse (such as women shouting at
their husbands) and psychological abuse (such as women shaming their husbands in
public). This finding was nationally representative, and was illustrated, for example, by
the following two respondents in the Gbawe mixed FGD (18–25):
Respondent 8: Women in times of violence are physically assaulted, but men are verbally
abused.
Respondent 5: Women are abused physically and verbally, but men are just verbally abused.
4.3. Perpetration of domestic violence in Ghana
The results from the survey, as discussed above, showed that the perpetration of domestic
violence varied considerably between men and women, with women being more likely
than men to have experienced violence perpetrated within domestic relations. This section
discusses in more detail the profiles of domestic violence perpetrators. The section starts
by analysing patterns of violence perpetration within domestic relations (subsection 4.3.1).
Subsection 4.3.2 discusses the profiles of study respondents who reported having
committed social, physical, sexual, psychological and/or economic violence against
another domestic relation. Subsection 4.3.3 focuses on the perpetration of violence against
children and young people.
4.3.1. Violence perpetration within domestic relations
Tables 43 and 44 provide an overview of the main domestic perpetrators of violence for
each type of violence – social, physical, sexual, psychological and economic – as reported
by female and male respondents with respect to the 12 months prior to the survey. It is
important to note that more than one perpetrator could perform different types of violence.
For simplicity, this table shows the aggregate figures of respondents’ first response.
151
Table 43: Perpetrators of domestic violence as reported by women (in percentages)
Panel A: Respondent is head of household (N=1,304)
Social Sexual Physical Psychological Economic
Partner/former partner 48.0 92.0 78.0 34.0 74.0
Spouse 0.0 0.0 5.0 3.0 2.0
In-laws 5.0 3.0 3.0 8.0 2.0
Child 2.0 0.0 0.0 0.0 2.0
Parent 12.0 0.0 1.0 11.0 8.0
Brother, sister 17.0 0.0 5.0 22.0 7.0
Uncle, aunt 2.0 3.0 1.0 6.0 4.0
Cousin, niece, nephew 4.0 0.0 6.0 8.0 0.0
Other relative 9.0 3.0 2.0 9.0 2.0
N 174 36 103 158 183
Panel B: Respondent is a spouse of the head of household (N=958)
Social Sexual Physical Psychological Economic
Head of household 63.0 96.0 81.0 47.0 83.0
Partner/former partner 1.0 2.0 0.0 4.0 2.0
Children 7.0 0.0 1.0 1.0 2.0
Parent of head 11.0 0.0 0.0 17.0 0.0
Parent 3.0 0.0 5.0 0.0 1.0
Brother, sister 4.0 0.0 4.0 5.0 6.0
Uncle, aunt 1.0 0.0 0.0 4.0 2.0
Cousin, niece, nephew 3.0 0.0 4.0 0.0 1.0
Other relative 8.0 0.0 4.0 17.0 2.0
N 123 24 75 78 122
Panel C: Respondent is the daughter/son of the head of household (N=246)
Social Sexual Physical Psychological Economic
Parent 65.0 18.0 22.0 51.0 50.0
Brother, sister 9.0 0.0 13.0 16.0 16.0
Partner of head 0.0 0.0 3.0 5.0 0.0
Partner/former partner 18.0 64.0 56.0 14.0 29.0
Uncle, aunt 3.0 0.0 0.0 9.0 3.0
Cousin, niece, nephew 3.0 0.0 6.0 2.0 0.0
Other relative 3.0 18.0 0.0 2.0 3.0
N 118 11 32 43 38
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Note: The totals N refer, respectively, to the number of respondents who were the head of the household, the
number of respondents who were spouses of the head of the household and the number of respondents who
were either the daughter or son of the head of the household.
The most predominant perpetrators of domestic violence against women were their
partners or former partners: the share of perpetration by (former) partners ranged between
34 per cent (psychological violence) and 92 per cent (sexual violence) against female
household heads (Table 43, Panel A). The share of husbands perpetrating violence against
their wives ranged between 47 per cent (psychological violence) and 96 per cent (sexual
152
violence) (Table 43, Panel B). Parents, in-laws and brothers and sisters were also identified
as perpetrators of domestic violence. Physical strength was one of the main factors used
by focus group discussion participants across all regions to describe the prevalence of
men’s physical violence against women in the household. In-depth interviews with key
community members confirmed this finding – for example, in the description offered by a
community leader working in the Eastern region, in the East Akim district:
“The difference between how women experience violence and that of men is that the men are
stronger and do harm the women when they beat them.”
Table 44: Perpetrators of domestic violence as reported by men (in percentages)
Panel A: Respondent is head of households (N=1,655)
Social Sexual Physical Psychological Economic
Partner/former partner 40.0 78.0 11.0 29.0 15.0
Spouse 21.0 2.0 1.0 7.0 9.0
In-laws 2.0 0.0 4.0 3.0 0.0
Child 3.0 0.0 2.0 0.0 3.0
Parent 8.0 0.0 18.0 15.0 28.0
Brother, sister 8.0 0.0 18.0 15.0 25.0
Uncle, aunt 2.0 7.0 10.0 10.0 7.0
Cousin, niece, nephew 5.0 0.0 14.0 7.0 5.0
Other relative 11.0 10.0 18.0 13.0 8.0
N 149 30 49 157 113
Panel B: Respondent is a spouse of the head of the household (N=5)
Social Sexual Physical Psychological Economic
0.0 0.0 0.0 0.0 0.0
Panel C: Respondent is the daughter/son of the head of household (N=219)
Social Sexual Physical Psychological Economic
Parent 56.0 0.0 42.0 62.0 84.0
Brother, sister 24.0 0.0 17.0 23.0 13.0
Partner/former partner 15.0 100.0 0.0 4.0 0.0
Uncle, aunt 3.0 0.0 8.0 0.0 0.0
Cousin, niece, nephew 0.0 0.0 0.0 0.0 3.0
Other relative 3.0 0.0 33.0 8.0 0.0
N 34 2 12 26 31
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Note: The totals N refer, respectively, to the number of respondents who were the head of the household, the
number of respondents who were spouses of the head of the household and the number of respondents who
were either the daughter or son of the head of the household.
Domestic violence against male respondents was also perpetrated largely by (former)
partners. Economic or physical violence against male heads of household was mainly
perpetrated by parents (28 and 18 per cent, respectively) and siblings (25 and 18 per cent,
respectively) (Table 44). Men who experienced violence committed by parents and siblings
were generally younger men who lived with their family in the same household.
153
4.3.2. Profile of domestic violence perpetrators
The qualitative data from all regions confirmed the range of perpetrators discussed above.
Within the domestic setting, the majority of the perpetrators were men, who committed
violence against their partners (generally women, in heterosexual relationships). As
discussed above, these forms of violence were predominantly economic and physical.
Complementing the quantitative findings shown in Tables 43 and 44, the qualitative
research revealed that domestic violence was both horizontal (between partners) and
vertical (across generations, between caregivers and children). Further, horizontal forms
of violence did not only flow one way, exerted by men against their female partners. The
focus group discussions and in-depth interviews discussed – sometimes at length,
particularly in the Eastern, Ashanti and Volta regions – how men were abused by women,
something that is also revealed in Tables 43 and 44.
Through careful facilitation of the focus group discussions, particularly in groups of
married men and of unmarried men and women, the study found that men also
experienced physical violence. This was not a nationally or regionally representative
finding, but it was a finding reported in urban areas, where men spoke about psychological
and verbal abuse from their female spouses. In the Greater Accra region, one male
participant explained to the rest of the mixed male and female group (18–25 years old) that
women can be stronger than men, and that they can be physically violent to their spouses,
just like men:
“The one I know the woman is much stronger than the man. So she does not obey the rules
of the man. If the man tries to talk she just beats the man up.”
The GFLHS 2015 collected information that allowed the study to explore further the
direction of the perpetration of violence between men and women. Table 45 shows the
share and number of respondents in the survey who reported having perpetrated acts of
domestic violence in the 12 months prior to the survey.
Table 45: Percentage and number of domestic violence perpetrators
Percentage of respondents reporting perpetration of domestic violence
Sex of
perpetrator
Social
violence
Sexual
violence
Physical
violence
Psychological
violence
Economic
violence
Any domestic
violence
Male 0.12 0.06 0.06 0.15 0.06 0.30
Female 0.12 0.01 0.07 0.15 0.03 0.27
Total 0.12 0.03 0.07 0.15 0.04 0.28
154
Number of respondents reporting perpetration of domestic violence
Sex of
perpetrator
Social
violence
Sexual
violence
Physical
violence
Psychological
violence
Economic
violence
Any domestic
violence
Male 222 120 110 277 97 563
Female 362 26 208 422 79 789
Total 584 146 318 699 176 1,352
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Table 45 reveals that 563 men (0.3 per cent of the sample) and 789 women (0.27 per cent of
the sample) reported having perpetrated at least one act of domestic violence. Men were
more likely to report perpetrating sexual or economic violence (0.06 per cent in both cases,
corresponding to 120 and 97 men, respectively). More women than men reported having
perpetrated social, physical or psychological violence. These results must be interpreted
with great care because they are likely to be vastly under-reported. The results may also
exhibit reporting biases. Given the mismatch between the results in Table 45 and the
qualitative data, it is possible that women may have been more prone to report
perpetration of violence than men. With these caveats in mind, it is nonetheless noteworthy
that violence is perpetrated by both sexes. Table 46 provides further disaggregation of the
above results.
Table 46: Number and percentage of perpetrators reporting violence against same or
another sex
Social violence
Sexual violence
Physical violence
Psychological violence
Economic violence
Any domestic violence
Female perpetrators against male or female household members
Against men 176 (0.62) 3 (1.0) 90 (0.60) 106 (0.69) 30 (0.59) 248 (0.67)
Against women 193 (0.69) - 92 (0.61) 93 (0.61) 33 (0.65) 233 (0.63)
Total number 281 3 150 153 51 370
Male perpetrators against male or female household members Against men 71 (0.55) - 24 (0.69) 31 (0.49) 12 (0.24) 84 (0.50)
Against women 95 (0.73) 9 (0.90) 17 (0.49) 48 (0.76) 43 (0.86) 126 (0.75)
Total number 130 10 35 63 50 169
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Table 46 shows that the largest share of domestic violence committed by women was
against men (0.67 per cent), but the difference between this and the share of violence
committed by women against other women (0.63 per cent) is small. Women were most
likely to commit violence against other women in the cases of social violence, physical
155
violence and economic violence (0.69, 0.61 and 0.65 per cent of women reported having
committed these forms of violence, respectively, against other women). Most of the
violence committed by men was against women (0.75 per cent, in comparison to 0.50 per
cent against other men). It is again important to keep in mind that these estimates may be
problematic, as discussed above. However, Tables 45 and 46 illustrate the level of
complexity and variety of domestic violence perpetration in Ghana.
Episodes of physical violence were also described as dynamic in that they could be driven
at first by one partner but ‘ended’ by another. This quote illustrates such dynamics, widely
reported at a regional level (but not nationally representative):
“The man gave the woman some blows on any part of her body that was available until she
fell down. The woman stood up and held the man by the neck, and the man began to give her
some blows until both of them fell down. They were so tired that they lay down till everybody
left the scene.” (FGD, Ashanti region, Adaase district: male and female group, aged
18+)
In some cases, the instance of physical violence followed a period of other forms of
violence. For example, some focus group discussions described how women struggled
with economic violence (such as men’s refusal to provide chop money) and that this violence
was compounded when women were ‘punished’ by their partner for failing to cook food
(itself a retaliation against the failure to provide money). Structural violence, and
pernicious poverty, might underlie this dynamic of intersecting forms of violence, as men
were not always able to secure wage labour and struggled with the pressure to ensure the
economic well-being of their household and its members (as is discussed later in the
report).
The qualitative data found, too, that some women withheld sex or shamed their partners
(through verbal abuse) when they did not earn enough money to support the running costs
of the household. This suggests an interlinked ‘web of violence’ that requires multiple
approaches for redress.
The study also collected information on how the perpetration of domestic violence varied
across socio-economic characteristics and geographical location of women and men. These
data are summarised in Tables 47 and 48, which suggest the following profiles:
The incidence of domestic violence perpetrated by women was highest among self-
employed and married women aged 25–29 years who lived in rural areas with no
or primary education and in the lowest asset quintile.
The incidence of domestic violence perpetrated by men was highest among men
aged 15–19 years who were employed and married, who lived in rural areas, were
in the lowest asset quintile but had primary or secondary education.
156
Table 47: Percentage of women reporting perpetration of domestic violence
Social Sexual Physical Psychological Economic Any
domestic
violence
Age group (years)
15–19
0.07
0.01
0.08
0.19
0.01
0.29
20–24 0.10 0.02 0.05 0.11 0.01 0.23 25–29 0.14 0.01 0.10 0.20 0.02 0.33 30–39 0.13 0.01 0.09 0.15 0.04 0.29 40–49 0.14 0.01 0.07 0.14 0.02 0.27 50–60 0.09 0.00 0.03 0.09 0.04 0.18
Employment
Self-employed
0.14
0.01
0.08
0.15
0.03
0.28 Employed 0.10 0.01 0.09 0.11 0.02 0.24 Not working 0.08 0.01 0.06 0.16 0.03 0.25
Marital status
Never married
0.04
0.00
0.09
0.17
0.02
0.25
Married or living together 0.12 0.01 0.08 0.14 0.03 0.27 Divorced/separated/widowed 0.13 0.01 0.05 0.15 0.03 0.26
Residence
Urban
0.11
0.01
0.07
0.13
0.02
0.25
Rural 0.13 0.00 0.08 0.17 0.03 0.29
Region
Western
0.14
0.01
0.09
0.16
0.04
0.29
Central 0.06 0.01 0.04 0.08 0.04 0.18 Greater Accra 0.10 0.02 0.07 0.11 0.03 0.21 Volta 0.09 0.01 0.11 0.31 0.02 0.39 Eastern 0.23 0.01 0.09 0.18 0.03 0.37 Ashanti 0.12 0.01 0.07 0.15 0.02 0.29 Brong Ahafo 0.08 0.00 0.06 0.11 0.02 0.20 Northern 0.14 0.00 0.13 0.18 0.03 0.33 Upper East 0.14 0.00 0.04 0.04 0.01 0.16 Upper West 0.15 0.04 0.04 0.08 0.02 0.20
Education level
None
0.13
0.01
0.08
0.15
0.03
0.28 Primary 0.14 0.00 0.08 0.16 0.04 0.28 Middle/JSS/JHS 0.12 0.01 0.07 0.15 0.02 0.27 Secondary 0.09 0.02 0.08 0.12 0.02 0.25 Technical 0.07 0.03 0.11 0.12 0.00 0.24 Higher 0.08 0.02 0.06 0.11 0.02 0.22
Asset quintile
Lowest
0.11
0.01
0.09
0.18
0.04
0.29
Second 0.14 0.01 0.07 0.17 0.03 0.28 Middle 0.10 0.01 0.07 0.16 0.02 0.27 Fourth 0.12 0.01 0.07 0.12 0.02 0.25 Highest 0.12 0.02 0.08 0.09 0.02 0.25
Total 0.12 0.01 0.07 0.15 0.03 0.27
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
157
Table 48: Percentage of men reporting perpetration of domestic violence
Social Sexual Physical Psychological Economic Any
domestic
violence
Age group (years)
15–19
0.07
0.06
0.09
0.22
0.04
0.35
20–24 0.12 0.12 0.07 0.18 0.04 0.32 25–29 0.12 0.10 0.07 0.14 0.04 0.33 30–39 0.10 0.04 0.05 0.14 0.07 0.27 40–49 0.16 0.04 0.04 0.12 0.09 0.33 50–60 0.11 0.02 0.04 0.14 0.04 0.25
Employment
Self-employed
0.12
0.05
0.05
0.13
0.05
0.28
Employed 0.12 0.07 0.07 0.15 0.08 0.34 Not working 0.11 0.08 0.06 0.19 0.03 0.32
Marital status
Never married
0.06
0.05
0.07
0.16
0.03
0.25
Married or living together 0.14 0.06 0.06 0.16 0.07 0.33 Divorced/separated/widowed 0.10 0.09 0.05 0.12 0.03 0.27
Residence
Urban
0.10
0.08
0.06
0.15
0.05
0.30
Rural 0.13 0.05 0.06 0.16 0.06 0.31
Region
Western
0.12
0.04
0.03
0.17
0.12
0.34
Central 0.09 0.09 0.08 0.13 0.03 0.26 Greater Accra 0.10 0.11 0.05 0.10 0.06 0.26 Volta 0.07 0.07 0.13 0.33 0.03 0.41 Eastern 0.24 0.05 0.08 0.17 0.05 0.40 Ashanti 0.13 0.07 0.05 0.16 0.04 0.33 Brong Ahafo 0.09 0.06 0.04 0.10 0.08 0.23 Northern 0.08 0.02 0.03 0.13 0.02 0.24 Upper East 0.13 0.00 0.07 0.05 0.05 0.23 Upper West 0.14 0.03 0.00 0.06 0.03 0.24
Education level
None
0.13
0.02
0.05
0.15
0.05
0.29
Primary 0.14 0.10 0.07 0.16 0.06 0.36 Middle/JSS/JHS 0.10 0.05 0.05 0.15 0.05 0.29 Secondary 0.10 0.10 0.07 0.16 0.05 0.32 Technical 0.09 0.08 0.07 0.15 0.05 0.25 Higher 0.14 0.05 0.05 0.15 0.08 0.31
Asset quintile
Lowest
0.10
0.04
0.06
0.16
0.06
0.31 Second 0.11 0.07 0.05 0.16 0.05 0.31 Middle 0.12 0.06 0.06 0.13 0.06 0.28 Fourth 0.13 0.07 0.07 0.16 0.06 0.31 Highest 0.13 0.08 0.07 0.14 0.05 0.30
Total 0.12 0.06 0.06 0.15 0.06 0.30
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
158
4.3.3. Domestic violence against children
In addition to the forms of domestic violence discussed in the sections above, the
qualitative data noted the occurrence of violence against children by adults who shared
the same home. The qualitative data, from all regions, found that women and men
employed physical violence to punish children, but there were different views on whether
physical violence, in particular, was an appropriate form of punishment for misbehaving
children. These views also differed depending on whether the person committing the
violence, or enacting the punishment, was male or female. For example, participants in a
focus group discussion with unmarried men and women (18–25 years old) in the Greater
Accra region said that only the father should beat children “because he is the breadwinner”.
In another focus group, also in the Greater Accra region, one respondent linked beating
with ‘correction’, saying, “fathers are responsible for correcting children through beating” (FGD,
Greater Accra region, Kukurantami district: male and female group, 18–25 years old).
Elsewhere, however, participants in the focus groups argued that this responsibility should
not be left only to men. In a focus group with married men (18–25 years old) in the Eastern
region, one participant asserted, “women should beat the children and the youth in the house
whenever they do something bad”. The GFLHS 2015 on domestic violence committed by
adults against their biological children found that the main perpetrator was the head of the
household (Table 49).
Table 49: Perpetration of domestic violence against individuals who reported domestic
violence and lived with their parents (in percentages)
Social
violence
Sexual
violence
Physical
violence
Psychological
violence
Economic
violence
Male head of household 88.0 0.0 100.0 87.0 82.0
Female head of household 87.0 100.0 63.0 70.0 87.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Note: Values were reported by the household respondent and refer to children of the household head.
The qualitative data also highlighted that non-biological children may be at particular risk
of domestic violence inflicted by stepfathers, particularly when the mother is in a
vulnerable economic position. The following statement reflects this widespread finding
across regions:
“[Domestic violence] is a multi-faceted problem involving couples, children and at times in-
laws as well. Yes, at times it involves adults and their non-biological children. You
sometimes hear of men making advances towards the daughter of his wife who was brought
into the marriage. You can’t believe this, but it is true. When the man is supposed to treat
that child as his own daughter, he rather decides to make amorous advances towards her.”
(IDI, Upper East region, Garu Tempane district assembly, Garu opinion leader)
159
In addition to physical violence, controlling behaviour, such as fathers refusing to let their
daughters attend school, was noted as a form of domestic violence across all regions in the
qualitative data. The focus group participants were in significant agreement that this form
of violence was unacceptable. Given the nature of the qualitative data, it was not possible,
as with the quantitative survey, to gauge the full extent of these different forms of violence.
However, the data offer important insights into some of the underlying social norms,
beliefs and attitudes that individuals, families and communities hold around certain forms
of violence and their acceptability. These findings are discussed in further detail in the
section below.
4.4. Attitudes towards domestic violence in Ghana
The second research question asked how attitudes and social norms may influence the level
of acceptance of domestic violence in Ghana. Attitudes and beliefs – even if held at a
personal level – are usually shaped by interactions with people within the family; with
people in the immediate environment, including communities, neighbourhoods, social
networks and workplaces; and within economic or political structures. Understanding
how these reproduce or challenge gender inequality, gender norms and domestic violence
is crucial for designing interventions and policies that will effectively decrease the
incidence of domestic violence. For this reason, this component of the study sought to
explore, first, how individuals perceived and experienced domestic violence in their own
lives and families (the individual and community in the social ecological model) and,
second, how perceptions and experiences of individuals were informed by social norms
that may be collectively reinforced at the level of the community and broader social
contexts.
During the quantitative data collection, men and women were asked to discuss five
different scenarios that involved acts of violent behaviour. The aim of this exercise was to
understand to what extent these violent acts were perceived as acceptable, to what extent
victims of domestic violence were blamed for the violence (and under which
circumstances), and what respondents felt would be acceptable punishments for different
acts of violence. This information was complemented with qualitative data from focus
group discussions and interviews. To generate data to complement the survey, the focus
groups were also asked to respond to and discuss similar scenarios of social, physical,
sexual, psychological and economic violence. The focus group discussions and in-depth
interviews with opinion leaders, activists and legal experts explored in addition whether
and why certain types of behaviour were considered acceptable.
An important finding from the research centred on the variation in people’s understanding
of what constitutes: (i) violence in general; and (ii) domestic violence in particular. As a
result, the study offers a set of nuanced findings that complicate a straightforward reading
of domestic violence as per the Domestic Violence Act. Importantly, these nuanced
findings suggest that successful interventions need to be carefully tailored to raise
160
awareness around each of the categories of violence that are included under the general
term of ‘domestic violence’.
Given the scale of the qualitative and quantitative research, this section has been structured
to reveal findings for each of the five main types of violence explored in the study – namely:
social, physical, sexual, psychological and economic. While data have been disaggregated
against these forms of violence, it is important to recognise that individuals can experience
multiple forms of violence simultaneously; wherever possible, the multiplicity of domestic
violence is highlighted in the discussion below.
4.4.1. Attitudes towards domestic social violence
Both the quantitative and qualitative data collection made use of vignette scenarios to
understand individual and group-level attitudes towards domestic social violence. The
GFLHS 2015 included a range of vignette scenarios, and survey respondents were
randomly assigned scenario questions. In doing so, the survey generated findings about
different attitudes towards domestic violence by sex and other socio-economic
characteristics of the respondents. One of the first scenarios intended to find out about the
acceptability of jealous behaviour by a girlfriend or a boyfriend. Examples of jealous
behaviour in the vignette included: checking messages on their partner’s phone; following
them around; showing up at their workplace unannounced; or demanding to know their
whereabouts all the time. A second vignette scenario assessed respondents’ attitudes
towards women who stalked men, and compared those to attitudes towards men who
stalked women. Results for this analysis are included in Tables 50 and 51, respectively.
Overall, the majority of respondents thought that neither men nor women should ever
resort to the two forms of controlling behaviour hypothesised in the scenario questions.
The results for the first scenario were very consistent (Table 50), with around 70 per cent
of respondents on average across all socio-economic groups and sex answering that ‘he
should never do that’. On average, around 20 per cent of the sample across all groups
answered ‘I sympathise with him, but he should not do this’. There were interesting
patterns among those who answered ‘Yes, he has the right to do this’. Fifteen per cent of
the sample overall answered the question in this way, but men and individuals aged 15–
19 years, living in rural areas, in the Upper East and Upper West regions, with no education
and in the lowest asset quintile were more likely than others to answer ‘yes’ to this question
(Table 50). The reactions to the second scenario question were very similar (Table 51).
161
Table 50: Responses to first social violence scenario (in percentages)
Yes, he has
the right to
do this
I sympathise with
him, but he should
not do that
He should never
do that
Total
Sex
Male 12.1 19.5 68.4 100.0 Female 10.9 19.7 69.4 100.0
Age group (years)
15–19 14.4 19.8 65.8 100.0 20–24 8.6 17.8 73.6 100.0 25–29 9.6 18.3 72.0 100.0 30–39 13.0 18.4 68.6 100.0 40–49 11.0 22.7 66.3 100.0 50–60 11.8 21.4 66.8 100.0
Employment
Self-employed 11.5 19.3 69.1 100.0 Employed 11.7 20.6 67.7 100.0 Not working 10.7 19.7 69.6 100.0
Marital status
Never married 9.4 22.0 68.6 100.0 Married or living together 11.5 19.2 69.3 100.0 Divorced/separated/widowed 11.9 20.2 67.8 100.0
Residence
Urban 9.9 19.2 70.9 100.0 Rural 13.0 20.1 66.9 100.0
162
Yes, he has
the right to
do this
I sympathise with
him, but he should
not do that
He should never
do that
Total
Region
Western 7.0 25.6 67.5 100.0 Central 12.2 30.9 56.9 100.0 Greater Accra 8.6 13.3 78.1 100.0 Volta 4.5 32.3 63.1 100.0 Eastern 14.5 15.9 69.6 100.0 Ashanti 10.7 21.5 67.8 100.0 Brong Ahafo 11.7 17.7 70.6 100.0 Northern 7.5 10.7 81.9 100.0 Upper East 32.3 14.0 53.7 100.0 Upper West 49.0 8.5 42.5 100.0
Education level
None 14.3 21.3 64.5 100.0 Primary 12.2 21.0 66.8 100.0 Middle/JSS/JHS 9.5 19.5 71.0 100.0 Secondary 12.9 16.5 70.6 100.0 Technical 5.3 12.6 82.1 100.0 Higher 10.4 21.5 68.1 100.0
Asset quintile
Lowest 14.6 21.9 63.5 100.0 Second 12.0 18.4 69.6 100.0 Middle 8.0 22.6 69.4 100.0 Fourth 10.5 17.8 71.7 100.0 Highest 11.1 18.5 70.4 100.0
Total 11.4 19.6 69.0 100.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
163
Table 51: Responses to second social violence scenario (in percentages)
Yes, she has
the right to
do this
I sympathise with
her, but she should
not do that
She should never
do that
Total
Sex
Male 9.7 18.9 71.4 100.0 Female 10.1 19.3 70.6 100.0
Age group (years)
15–19 11.6 19.5 68.8 100.0 20–24 12.3 19.7 67.9 100.0 25–29 11.2 16.9 72.0 100.0 30–39 10.1 17.2 72.6 100.0 40–49 6.8 22.6 70.6 100.0 50–60 8.2 20.9 71.0 100.0
Employment
Self-employed 9.3 19.2 71.5 100.0 Employed 8.8 18.0 73.2 100.0 Not working 12.4 19.8 67.8 100.0
Marital status
Never married 15.1 15.9 69.1 100.0 Married or living together 9.1 20.3 70.6 100.0 Divorced/separated/widowed 10.2 16.9 73.0 100.0
Residence
Urban 9.7 18.9 71.4 100.0 Rural 10.2 19.4 70.4 100.0
164
Yes, she has
the right to
do this
I sympathise with
her, but she should
not do that
She should never do that
Total
Region
Western 9.4 21.9 68.7 100.0 Central 12.9 27.9 59.2 100.0 Greater Accra 9.5 18.5 72.1 100.0 Volta 1.1 29.3 69.6 100.0 Eastern 6.0 14.7 79.3 100.0 Ashanti 10.1 16.7 73.2 100.0 Brong Ahafo 11.1 14.9 74.0 100.0 Northern 12.6 14.9 72.5 100.0 Upper East 21.5 10.5 67.9 100.0 Upper West 28.4 20.9 50.7 100.0
Education level
None 12.6 17.2 70.2 100.0 Primary 9.4 18.4 72.3 100.0 Middle/JSS/JHS 8.6 19.2 72.2 100.0 Secondary 11.7 20.6 67.8 100.0 Technical 7.8 19.2 73.0 100.0 Higher 8.4 22.1 69.6 100.0
Asset quintile
Lowest 7.9 21.2 70.9 100.0 Second 11.4 17.8 70.8 100.0 Middle 12.4 20.1 67.5 100.0 Fourth 10.4 17.7 71.8 100.0 Highest 6.9 19.5 73.5 100.0
Total 9.9 19.1 70.9 100.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
165
As noted above, the qualitative research also included scenarios to understand the
participants’ attitudes towards social violence. These scenarios were discussed in all 20
focus groups. The findings from this component of the research were nationally
representative. They are discussed in detail below in relation to the quantitative findings,
and specific quotes are used selectively to illustrate certain findings.
Two complementary vignette scenarios on attitudes towards social violence were
discussed in the focus group discussions, whereby participants were asked to discuss: (i)
whether it was acceptable for men to prevent women from working (outside the home);
and (ii) whether it was acceptable for men to prevent their daughters from attending
school. The qualitative findings show that in most regions (with the exception of the three
northern regions) respondents found that it was unacceptable for men to prevent their
wives from working – particularly when the women contributed extra income to the
household. Results for all regions also showed widespread disapproval of men preventing
their daughters from attending school. Focus group discussions in the three northern
regions indicated higher variation in the response to whether or not it was acceptable for
men to prevent their partners from working. This finding is in line with a recently
published study on gender inequality in the Northern region (Alhassan and Odame, 2015),
where the authors outlined a set of household and contextual factors that worked against
government efforts to promote gender equality. These factors included limited availability
of household resources and poverty. Further socio-cultural practices, such as polygyny,
fostering, child marriage and boy-child preference were also shown to collectively
undermine women’s status relative to men, and to reinforce their economic dependence
on men in shared households. In the focus group discussions in the Northern region,
preventing women from working was viewed as acceptable because their work outside the
house would be at the cost of their domestic chores. The following quote illustrates this
view:
“[F]rankly speaking, juggling motherhood and wifely duties with a profession is no mean
joke. It’s really difficult. So if the man thinks his wife is suffering too much and tells her to
stay at home and take care of the house whiles he works to support them, I don’t see any
problems with that.” (FGD, Northern region, Sagnerigu district: female group, 18–50
years)
In contrast, the following dialogue, drawn from a focus group discussion with married
men (25–50 years old) in the Volta region, is an example of the prevailing response across
all regions to the scenario asking whether it is acceptable for men to prevent their
daughters from attending school:
Interviewer: Is it acceptable for fathers to stop daughters from attending school?
Respondent 7: It is not right for the father to deny his girl child education. Women are now
doing what some men cannot do. There are women who are taking care of their families.
Respondent 6: It is very wrong for the father to stop his girl child from attending school.
Women also need education to survive just like men.
Respondent 5: It is wrong to deny a girl child her right to education. Your daughter may
marry an educated person. The problem comes when the husband marries another woman
166
and the second wife happens to be educated, the man will definitely prefer the educated
woman to your daughter.
Respondent 4: It is not good for a man to ask his daughter to stop schooling. In this modern
era, women are equally doing what men can do.
Respondent 10: It is not right for a father to deny his girl child her right to education. The
reason is that capability is not based on gender.
Combined, the qualitative and quantitative findings drawn from the vignette scenarios
suggest that social violence was consistently viewed as unacceptable. However, social
violence between partners may also be shaped by gender norms – prevalent in the northern
regions – in which women were discouraged from engaging in paid labour. All
respondents consistently argued that it was unacceptable to prevent girls from acquiring
an education.
4.4.2. Attitudes towards domestic physical violence
As noted above, an important finding from the qualitative and quantitative data relates to
the variation in people’s understanding of what constitutes domestic violence. This section
offers a detailed analysis of attitudes towards domestic physical violence. The research
team again used scenario vignettes to elicit attitudes towards physical violence. The survey
introduced the following scenario: ‘After a marital dispute the wife refused to talk to her
husband, and refused to make his dinner and any other household chores. The husband
hit his wife, but without leaving a mark.’ The responses to this scenario are provided in
Table 52. Seventy-five per cent of male and female respondents found it completely
unacceptable for a man to beat his wife because she refused to talk to him and do
household chores after their dispute. In line with previous findings, this answer was most
common among respondents who were employed (82.3 per cent), living in urban areas
(80.2 per cent) and in the Greater Accra region (89.9 per cent), with higher education (83.8
per cent) and in the highest asset quintile (82.6 per cent).
167
Table 52: Responses to first physical violence scenario (in percentages)
Completely
unacceptable
Somewhat
unacceptable
Somewhat
acceptable
Completely
acceptable
Do not
know
Total
Sex
Male 75.0 13.4 4.8 6.4 0.4 100.0 Female 75.1 10.6 4.9 9.2 0.2 100.0
Age group (years)
15–19 74.3 7.7 5.6 11.0 1.3 100.0 20–24 70.5 14.7 4.3 9.7 0.7 100.0 25–29 77.4 10.1 4.9 7.6 0.0 100.0 30–39 75.5 11.9 4.9 7.6 0.0 100.0 40–49 75.5 10.4 5.5 8.3 0.3 100.0 50–60 75.3 14.2 4.0 6.2 0.3 100.0
Employment
Self-employed 72.1 13.1 5.1 9.5 0.1 100.0 Employed 82.3 7.2 4.0 6.0 0.5 100.0 Not working 77.4 11.2 4.8 5.9 0.6 100.0
Marital status
Never married 77.4 10.7 2.9 7.8 1.1 100.0 Married or living together 73.9 11.9 5.4 8.7 0.1 100.0 Divorced/separated/widowed 78.1 11.5 4.0 6.3 0.2 100.0
Residence
Urban 80.2 9.9 3.5 6.0 0.3 100.0 Rural 69.7 13.5 6.3 10.2 0.3 100.0
168
Completely
unacceptable
Somewhat
unacceptable
Somewhat
acceptable
Completely
acceptable Do not
know
Total
Region
Western 86.6 6.2 3.3 4.0 0.0 100.0 Central 64.8 18.0 7.9 8.4 0.9 100.0 Greater Accra 89.9 5.5 3.1 1.2 0.3 100.0 Volta 69.9 22.8 2.6 4.0 0.7 100.0 Eastern 82.3 10.4 1.8 5.5 0.0 100.0 Ashanti 82.6 6.6 3.8 6.6 0.5 100.0 Brong Ahafo 71.7 12.1 2.6 13.6 0.0 100.0 Northern 44.5 22.2 13.9 19.4 0.0 100.0 Upper East 54.0 17.7 8.1 20.2 0.0 100.0 Upper West 32.0 19.9 16.7 31.5 0.0 100.0
Education level
None 58.6 17.5 8.6 15.3 0.0 100.0 Primary 70.8 11.7 5.3 11.2 1.0 100.0 Middle/JSS/JHS 80.0 9.9 4.1 5.7 0.4 100.0 Secondary 81.9 9.1 3.6 5.4 0.0 100.0 Technical 80.7 13.6 1.5 4.2 0.0 100.0 Higher 83.8 10.2 2.5 3.4 0.0 100.0
Asset quintile
Lowest 62.8 16.2 6.5 14.0 0.5 100.0 Second 73.3 13.0 6.4 7.2 0.0 100.0 Middle 76.6 10.9 3.7 8.5 0.3 100.0 Fourth 82.3 8.0 3.4 6.2 0.2 100.0 Highest 82.6 9.3 3.3 4.1 0.6 100.0
Total 75.1 11.7 4.9 8.1 0.3 100.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
169
Table 53: Responses to second physical violence scenario (in percentages)
Yes, fully Yes, partly Not at all Do not know Total
Sex
Male 48.2 32.7 18.5 0.6 100.0 Female 50.9 30.3 18.5 0.3 100.0
Age group (years)
15–19 55.9 20.5 22.3 1.3 100.0 20–24 49.6 30.5 19.2 0.7 100.0 25–29 49.9 29.3 20.6 0.3 100.0 30–39 50.3 30.9 18.6 0.2 100.0 40–49 47.4 33.4 18.8 0.4 100.0 50–60 48.4 38.8 12.6 0.2 100.0
Employment
Self-employed 52.4 30.6 16.8 0.3 100.0 Employed 40.9 36.4 21.8 0.9 100.0 Not working 49.7 29.3 20.5 0.4 100.0
Marital status
Never married 47.2 30.4 21.4 1.1 100.0 Married or living together 50.3 30.8 18.5 0.4 100.0 Divorced/separated/widowed 49.5 33.1 17.4 0.0 100.0
Residence
Urban 43.8 36.7 19.0 0.4 100.0 Rural 56.1 25.5 18.0 0.4 100.0
170
Yes, fully Yes, partly Not at all Do not know Total
Region
Western 57.6 22.6 19.8 0.0 100.0 Central 50.1 40.1 8.9 0.8 100.0 Greater Accra 25.9 52.6 21.2 0.3 100.0 Volta 39.6 36.4 23.2 0.8 100.0 Eastern 51.1 30.4 18.1 0.4 100.0 Ashanti 56.2 24.0 19.3 0.5 100.0 Brong Ahafo 56.5 21.5 21.5 0.5 100.0 Northern 64.4 22.0 13.7 0.0 100.0 Upper East 62.7 23.3 13.3 0.7 100.0 Upper West 67.1 14.1 18.7 0.0 100.0
Education level
None 59.0 27.3 13.7 0.0 100.0 Primary 54.4 28.2 16.5 0.9 100.0 Middle/JSS/JHS 49.8 31.4 18.1 0.6 100.0 Secondary 43.9 30.6 25.5 0.0 100.0 Technical 34.8 47.5 17.7 0.0 100.0 Higher 38.0 39.3 22.7 0.0 100.0
Asset quintile
Lowest 52.6 28.7 18.1 0.6 100.0 Second 58.2 25.6 16.1 0.2 100.0 Middle 51.4 26.9 21.4 0.4 100.0 Fourth 43.4 36.3 19.9 0.4 100.0 Highest 39.5 41.5 18.4 0.6 100.0
Total 49.8 31.2 18.5 0.4 100.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
171
The qualitative data from all regions similarly found that participants considered wife-
beating to be unacceptable. Wife-beating linked to women’s refusal to have sex was less
frequently mentioned in the focus group discussions, but when it was mentioned, both
female and male respondents iterated a consistent attitude, condemning men’s treatment
of women ‘as slaves’. In the excerpt below, for example, married men (aged 18–50) in the
Volta region explained why they believed that physical violence against women was
unacceptable. Their views were representative of the region’s findings (these issues were
not picked up in other focus group discussions), and they reflect the national-level data
from the interviews and other regional focus groups, condemning men as ‘weak’ for
beating women:
Respondent 2: No, it’s not acceptable because she’s his wife and deserves to be treated with
respect. She’s not a slave.
Respondent 7: No, it’s not acceptable. This is because the bible entrusts us to treat each
other with respect and love. After marriage, the two individuals become one, and so the man
ought to treat her as he would like to be treated.
Respondent 10: No, I also think it’s not acceptable because conflicts and disagreements are
integral parts of every relationship and so we can’t fall on beating to drive home our points.
Beating usually doesn’t solve anything so we must employ other means to get our wives to
agree with us.
Analysis of the qualitative data indicated that views about physical violence differed
depending on the age and relationship of the two (or more) people involved in the violent
incident or series of interactions. The data (in all regions) revealed in particular that
physical violence by parents against children was appropriate, and sometimes even
necessary. This statement, by a participant in a focus group discussion with married men
(aged 18–50) in the Brong Ahafo region, indicates this distinction about the acceptability of
violence against children by adults, on the one hand, and between men and women, on the
other hand:
Respondent 4: No, there is no justification whatsoever to beat your wife, because it is not a
father and child relationship, but you’re coming together as friends, and so whatever the case
may be, both of them should be able to iron their differences peacefully.
In research on attitudes to violence, predominantly in horizontal relationships (between
partners), scholars have found that the perceived culpability of the victim (i.e. whether
they are perceived to be at fault) plays a significant role in men and women’s perceptions
of the violence (Garcia-Moreno et al., 2005). Seeking to explore this factor in relation to
physical violence, the survey asked whether the respondents thought it was the woman’s
fault when she was beaten by her husband – in relation to the scenario discussed above.
The results of this analysis were striking and are presented in Table 53. Around half of the
sample on average across all groups answered that they fully agreed that the beating was
the woman’s fault – even though the majority of the respondents found the husband’s
behaviour unacceptable. This answer was more common among women: 50.9 per cent of
all women agreed that the woman in the scenario was at fault, compared to 48.2 per cent
of male respondents. The answer was also more common among self-employed
172
individuals (52.4 per cent), married (50.3 per cent), living in rural areas (56.1 per cent), with
no education (59.0 per cent) and in the lowest two asset quintiles (52.6 and 58.2 per cent,
respectively) (Table 53).
The survey also asked respondents under what circumstances wife-beating would be
acceptable. The five most accepted reasons for wife-beating are shown in Table 54, while
Table 55 expands this analysis across different socio-economic groups. The main reasons
why men and women justified wife-beating included disobedience and neglect of children:
11.2 per cent of male respondents and 17.2 per cent of female respondents mentioned that
wife-beating was acceptable if the wife had disobeyed the husband, and 10.2 per cent of
male respondents and 16.5 per cent of female respondents agreed that wife-beating was
acceptable if the wife neglected the children (Table 54). If the woman refused sex, a form
of violence largely condemned in all focus group discussions where this issue was raised,
5.7 per cent of men and 9.8 per cent of women thought this was an acceptable reason for
wife-beating (Table 55, Column 3).
Table 54: Percentage of respondents agreeing with wife-beating if the wife…
Male
respondents
Female
respondents
…disobeys her husband 11.2 17.2
…neglects the children 10.2 16.5
…goes out without telling her husband 7.3 13.2
…contracts a sexually transmitted disease 7.8 13.2
…argues with her husband 7.0 11.8
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
As noted above, women were consistently more likely to find wife-beating acceptable than
men (Tables 54 and 55). This finding echoes a similar result discussed in Takyi and Mann
(2009) and may imply that victims internalise social norms once they are in abusive
relations, as discussed in Section 2.
In line with findings discussed above, Table 53 shows that condoning wife-beating was
also more common, on average, among younger individuals, either self-employed or not
working, married, living in rural areas, with no education and in the bottom income
quintiles. There were also large regional differences: respondents from the three northern
regions were more likely to reveal attitudes supporting wife-beating than other regions.
173
Table 55: Percentage of respondents who think that a man has a good reason to hit his wife if:
1 2 3 4 5 6 7 8 9 10 11
Sex Male 4.6 11.2 5.7 2.0 5.6 0.4 7.8 7.3 10.2 7.0 2.4 Female 7.7 17.2 9.8 3.2 8.7 0.5 13.2 13.2 16.5 11.8 4.3
Age group
15–19 10.6
21.7 11.7 2.8 9.6 1.6 15.5 17.0 18.4 13.7 4.9 20–24 5.9 15.3 7.8 4.0 9.4 0.4 12.1 10.9 16.5 9.4 5.1 25–29 5.5 14.5 7.7 2.3 6.0 0.3 10.1 10.0 12.9 10.3 2.8 30–39 5.5 12.9 6.4 2.2 6.3 0.3 10.2 9.2 12.7 8.5 3.2 40–49 6.4 15.3 10.0 3.0 9.1 0.3 11.9 12.0 14.9 10.6 3.8 50–60 7.6 13.7 8.4 2.8 6.8 0.5 9.5 10.1 12.0 9.5 2.8
Employment
Self-employed 7.0 16.4 9.2 3.3 8.2 0.4 12.5 11.6 15.5 11.1 3.9 Employed 3.2 6.8 4.6 0.9 4.2 0.4 5.7 6.0 8.7 4.9 2.0 Not working 7.4 16.5 8.1 2.6 8.0 0.7 11.2 12.3 14.0 10.2 3.8
Marital status Never married 5.9 12.8 6.6 1.4 6.1 0.9 9.6 8.6 10.0 7.2 2.6 Married or living together 6.6 15.5 8.6 3.1 7.9 0.3 11.4 11.2 14.3 10.5 3.7 Divorced/separated/widowed 6.0 13.2 7.6 2.1 6.7 0.6 10.6 10.5 15.0 8.9 3.6
Residence Urban 3.4 9.3 4.8 1.3 4.7 0.4 7.2 6.9 9.9 6.0 1.6 Rural 9.6 20.5 11.7 4.2 10.5 0.5 15.1 15.0 18.3 13.9 5.6
174
1 2 3 4 5 6 7 8 9 10 11
Region
Western 4.4 14.1 8.2 0.9 10.0 0.2 8.9 13.0 13.2 8.3 4.0 Central 5.9 16.6 8.1 0.5 4.7 0.6 13.3 9.4 10.5 4.9 1.3 Greater Accra 1.2 3.1 1.7 1.0 2.6 0.1 3.0 2.6 3.9 1.7 0.8 Volta 4.9 7.3 5.2 3.3 4.7 1.6 9.9 7.2 9.5 8.0 4.5 Eastern 5.3 11.8 7.0 1.6 3.6 0.3 15.4 7.7 18.2 8.6 2.4 Ashanti 5.7 12.4 4.8 1.0 6.5 0.3 6.1 8.5 9.5 7.3 2.3 Brong Ahafo 3.6 16.1 7.3 2.7 7.7 0.5 10.1 8.6 12.4 12.8 2.6 Northern 14.8 42.7 24.4 11.3 15.0 0.6 21.3 28.5 38.6 29.1 6.7 Upper East 25.8 32.4 24.0 10.8 26.4 0.4 37.4 36.9 36.4 27.0 17.1 Upper West 31.5 38.6 29.4 8.9 27.8 0.4 34.9 30.2 37.4 29.3 16.7
Education level None 12.6 28.6 17.5 7.0 13.5 0.4 19.9 21.9 25.5 20.9 7.6 Primary 8.1 17.7 9.9 3.8 9.4 0.9 14.6 12.3 17.0 12.4 4.6 Middle/JSS/JHS 5.5 12.6 6.3 1.7 6.4 0.4 9.5 8.9 12.2 7.9 3.1 Secondary 3.1 8.2 3.5 0.9 4.9 0.3 6.2 6.7 8.5 4.8 1.2 Technical 2.0 4.3 5.1 0.6 4.3 0.1 5.2 2.6 3.9 1.3 0.7 Higher 1.6 4.6 2.5 0.3 1.8 0.1 3.4 2.9 5.1 2.5 0.5
Asset quintile
Lowest 10.6 22.5 13.4 4.7 11.2 0.9 18.5 17.7 20.5 15.5 6.0 Second 8.2 18.7 11.0 4.2 8.9 0.4 13.3 13.4 17.3 12.6 5.3 Middle 5.0 13.8 6.8 1.4 7.8 0.2 8.8 10.1 12.5 9.6 2.5 Fourth 4.4 10.1 4.6 1.7 5.6 0.3 7.8 6.7 10.6 5.8 2.1 Highest 2.4 5.9 3.1 0.7 2.5 0.4 4.6 4.1 6.4 4.0 0.7
Total 6.4 14.8 8.2 2.7 7.5 0.4 11.1 10.8 14.0 9.9 3.6
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Note: 1 = She does not complete her household work to his satisfaction; 2 = She disobeys him; 3 = She refuses to have sex with him; 4 = She asks him whether
he has other girlfriends; 5 = He suspects that she is unfaithful; 6 = He finds out that she has been unfaithful; 7 = She contracts a sexually transmitted disease; 8
= She goes out without telling him; 9 = She neglects the children; 10 = She argues with him; 11 = She burns the food.
175
Fewer women in 2015 found wife-beating acceptable than in the 2008 DHS. Table 56
includes comparable questions asked in the GFLHS 2015 and in the Ghana DHS 2008. The
results show a reduction of 13.5 percentage points in the number of women justifying wife-
beating (from 36.6 per cent in 2008 to 23.1 per cent in 2015), and 8 percentage points in the
number of men condoning wife-beating for at least one of the reasons provided (from 21.8
per cent in 2008 to 13.8 per cent in 2015). Even though this attitude is still more prevalent
among women, it also seems to be changing more rapidly among women (the reduction is
larger among women).
Table 56: Percentage of respondents stating that wife-beating is acceptable if the
woman…
…refuses
to have sex
… goes out
without
asking
… neglects
children
… argues
with the
husband
… burns
food
At least
one
reason
2015 GFLHS
Men 5.7 7.3 10.2 7.0 2.4 13.8
Women 9.8 13.2 16.5 11.8 4.3 23.1
2008 DHS
Men 6.6 10.7 13.7 11.0 4.6 21.8
Women 12.3 22.3 26.3 21.3 8.3 36.6
Source: Ghana Family Life and Health Survey (GFLHS) 2015 and DHS 2008.
While the GFLHS 2015 focused predominantly on wife-beating, the qualitative focus group
discussions were able to explore in more detail men’s experiences of physical violence by
women – even though few numbers of respondents considered this a form of domestic
violence. This finding was regionally representative, with respondents in the Central,
Eastern, Volta and Greater Accra regions iterating the importance of ‘fairness’: just as
women should not be beaten, so too should women not beat men.
However, in the Northern and Upper East regions, the focus group discussions about
‘husband-beating’ – particularly among young men and women (aged 18–25) and among
married women (aged 18–50) – elicited a great deal of laughter, and sometimes even
disbelief. These responses to some of the group members’ accounts of men experiencing
physical violence were overlaid with gendered assumptions that these men were ‘weak’
and not able to maintain their superiority over women. For example, the excerpt below
from the Gbolo district in the Northern region offers a window into one such discussion,
where the possibility of a man being beaten by his wife was met with disbelief:
Respondent 2: If it is the woman beating the man, I would not separate them. I will pass by
because why should a man allow himself to be beaten by a woman?
Respondent 7: I will go and separate them, but if I could, I will support the man to sack the
woman/wife because it is disrespectful. Why should you beat up your husband?
Respondent 6: I will separate them.
176
Respondent 4: I will not mind them. I will pass by because why a woman should beat her
husband is a taboo in our community.
(FGD, Northern region, Gbolo district: male and female group, 18–25 years)
However, sometimes the laughter had a serious undertone, revealing discontent with
gender norms among the respondents, particularly in the Northern and Upper East
regions. In some groups, the question revealed a desire for revenge by women who had
themselves experienced multiple forms of domestic violence. In this case, this finding
suggests that violence may be deemed ‘acceptable’ when it constitutes, or is viewed to
constitute, a form of retribution. This sentiment, while widely reported mostly in the
Northern and Upper East regions, is powerfully captured in this discussion among
married women (aged 18–50).
Respondent 1: I’ll stand by and watch as the woman beats him, I’ll neither call for help nor
try to separate them because he’s beaten his wife and so he should also have a taste of what
it takes to be beaten.
Respondent 2: If I meet such a scenario, I’ll simply let the woman beat the man because he’s
a wife-beater and so he must be taught a lesson in respecting women, so I’ll cheer the woman
on.
Respondent 7: As for me, I’ll help the lady to give the man a good beating. I’ll even referee
the fight.
(FGD, Northern region, Sagnerigu district: female group, 18–50 years)
This dynamic, however, did not apply to men’s views of the acceptability of violence
against women.
The qualitative and quantitative data have also allowed the research team to examine
attitudes towards physical violence against children. This form of violence was
predominantly considered unacceptable. However, some study participants expressed
certain contexts when physical violence against children was acceptable or even desirable.
The qualitative data revealed how, across all regions, parents often struggled to discipline
their children when they misbehaved. Focus group discussions in all ten regions revealed
that both older and younger respondents of both sexes agreed that there were certain
circumstances when physical violence against children was acceptable. This finding
suggests that not only do gender norms shape the perceived acceptability of certain forms
of violence between partners, but that age is an important feature for distinguishing what
kinds of physical violence are and are not appropriate. This finding is poignantly
illustrated in the following excerpt from a focus group discussion in the Northern region,
which represents views expressed across the whole country:
Respondent 2: Yes, it is acceptable somehow because if they do what is not right, the father
has to talk to them and if they persist, then the father can beat them.
Respondent 4: Yes, it is acceptable somehow because the man has to talk to the children for
the third time and if they persist then the man can beat them.
Respondent 7: Yes, but it is very necessary for the men to talk to the children for the first
time and if they prove stubborn, then they have to beat them.
177
(FGD, Northern region, Gbolo district: male group, aged 50+)
To explore the circumstances under which physical violence against children was deemed
acceptable, the GFLHS 2015 included a scenario question asking respondents whether they
would intervene if they witnessed an act of physical violence against a child. Ninety per
cent of respondents said they would intervene through a variety of means, including their
own direct intervention or by approaching religious authorities, community leaders, a
relative of the child, a neighbour or other community members. 53 per cent of men and
50.8 per cent of women in the GFLHS 2015 mentioned that they would intervene directly
by going to check what happened, trying to rescue the child themselves or trying to stop
the dispute.
Physical punishments of children seemed to be at least partially acceptable if they intended
to discipline the child when non-physical attempts were ineffective. The GFLHS 2015
asked respondents whether they fully agreed, somewhat agreed, somewhat disagreed or
fully disagreed with the following statement: ‘To bring up, raise or educate a child
properly, the child sometimes needs to be physically punished.’ 50 per cent of women and
43.3 per cent of men fully agreed with this statement (Table 57). Younger respondents were
more likely to fully disagree with the physical punishment of children (over 24 per cent
among those aged 15–24 years, compared to 19.9 per cent among respondents aged 50–60
years), as were employed respondents, living in urban areas and in the Western and
Ashanti regions, and those with higher levels of education and assets (Table 57). There has
been, however, a significant change in attitudes towards physical violence against
children: in 1998, 8 per cent of respondents thought it was never acceptable to beat a child
(Coker-Appiah and Cusack, 1999). In 2015, this number has risen to 25.3 per cent among
men and 18.8 per cent among women (Table 57).
178
Table 57: Views about statement ‘To bring up, raise or educate a child properly, the
child sometimes needs to be physically punished’ (in percentages)
Fully
agree
Somewhat
agree
Somewhat
disagree
Fully
disagree
Do not
know
Total
Sex Male 43.3 18.9 12.3 25.3 0.1 100.0 Female 50.6 18.7 11.9 18.8 0.0 100.0
Age group (years) 15–19 48.1 15.2 11.8 24.8 0.0 100.0 20–24 45.7 17.6 12.3 24.3 0.0 100.0 25–29 47.6 19.7 10.8 22.0 0.0 100.0 30–39 48.1 18.3 12.4 21.1 0.1 100.0 40–49 49.6 18.9 12.7 18.7 0.1 100.0 50–60 46.4 21.4 12.2 19.9 0.0 100.0
Employment
Self-employed 49.2 19.1 12.7 19.1 0.0 100.0 Employed 44.2 17.8 12.2 25.5 0.3 100.0 Not working 46.4 18.8 10.4 24.4 0.0 100.0 Marital status
Never married 50.8 15.6 11.6 22.1 0.0 100.0 Married or living together 47.0 19.2 12.1 21.7 0.0 100.0 Divorced/separated/widowed 48.5 19.0 12.3 20.2 0.0 100.0 Residence
Urban 45.8 19.5 11.4 23.2 0.0 100.0 Rural 49.7 18.0 12.8 19.5 0.1 100.0 Region
Western 50.6 10.9 8.5 30.0 0.0 100.0 Central 46.5 17.7 22.1 13.7 0.0 100.0 Greater Accra 49.2 20.8 10.3 19.8 0.0 100.0 Volta 40.7 32.6 13.4 13.3 0.0 100.0 Eastern 67.2 18.2 6.7 7.9 0.0 100.0 Ashanti 45.4 8.8 8.2 37.3 0.2 100.0 Brong Ahafo 42.1 23.0 13.3 21.6 0.0 100.0 Northern 31.2 28.0 24.7 16.2 0.0 100.0 Upper East 54.2 18.7 10.1 17.1 0.0 100.0 Upper West 63.8 21.8 6.8 7.6 0.0 100.0 Education level
None 45.2 23.3 15.3 16.2 0.0 100.0 Primary 53.7 18.0 11.5 16.8 0.0 100.0 Middle/JSS/JHS 49.2 17.0 12.0 21.6 0.1 100.0 Secondary 44.8 17.3 9.5 28.5 0.0 100.0 Technical 38.6 25.5 12.1 23.8 0.0 100.0 Higher 45.0 18.4 11.0 25.7 0.0 100.0 Asset quintile
Lowest 45.3 21.9 14.1 18.8 0.0 100.0 Second 50.3 18.1 13.4 18.3 0.0 100.0 Middle 48.6 16.7 12.0 22.7 0.0 100.0 Fourth 47.4 18.3 10.7 23.4 0.1 100.0 Highest 45.8 19.0 9.2 25.9 0.1 100.0
Total 47.7 18.8 12.1 21.4 0.0 100.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
179
4.4.3. Attitudes towards domestic sexual violence
The GFLHS 2015 revealed nuanced attitudes of different individuals and social groups
towards domestic sexual violence, with many exhibiting a higher degree of acceptance of
some dimensions of sexual violence than other forms of violence. The research team
explored attitudes towards domestic sexual violence by asking respondents their views in
relation to two statements: (1) ‘If a woman wears revealing clothes, it is okay for men to
make some sexual comments as she walks in her neighbourhood’; and (2) ‘If a woman
wears revealing clothes, it is her fault if she is raped.’ Tables 58 and 59 show, respectively,
the reactions of the respondents to these two statements.
Findings show that a very substantial part of the sample – 36.7 per cent of men and 40.7
per cent of women – fully agreed with the first statement (Table 58), without much
variation in responses across socio-economic groups. Individuals with higher levels of
education, in the top asset quintile and living in the Eastern and Ashanti regions and in
urban areas were, however, more likely to fully disagree with the first statement (Table 58,
Column 4), even though ‘fully agree’ was the most common answer across most socio-
economic groups.
More than three in every five respondents (61.6 per cent) fully agreed that the woman was
to be blamed for rape if she wore revealing clothes. As with other forms of violence, a
higher proportion of women than men displayed attitudes condoning this: 65.3 per cent of
women interviewed agreed with the second statement, compared to 56.2 per cent of men.
Respondents from rural areas were more likely than those in urban areas to fully agree
with the second statement (65.8 and 57.7 per cent, respectively). The data show no other
substantial differences across other group divisions. Employed individuals with higher
levels of education, in the top asset quintile and living in urban areas were more likely to
fully disagree with the second statement (but ‘fully agree’ was the most common answer
among most socio-economic categories).
180
Table 58: Views about statement ‘If a woman wears revealing clothes, it is okay for
men to make some sexual comments as she walks in her neighbourhood’ (in
percentages)
Fully
agree
Somewhat
agree
Somewhat
disagree
Fully
disagree
Do not
know
Total
Sex
Male 36.7 17.2 16.6 28.9 0.6 100.0
Female 40.7 16.9 14.4 27.4 0.6 100.0 Age group (years)
15–19 40.4 15.0 13.6 28.6 2.4 100.0 20–24 40.6 14.8 13.7 30.8 0.1 100.0 25–29 38.9 17.4 14.9 28.5 0.3 100.0 30–39 38.2 17.5 16.3 27.5 0.5 100.0 40–49 40.8 17.4 15.1 26.0 0.6 100.0 50–60 36.9 18.3 16.4 27.7 0.7 100.0
Employment
Self-employed 39.5 17.9 15.9 26.2 0.5 100.0 Employed 38.3 15.7 14.2 31.4 0.5 100.0 Not working 38.5 15.9 14.5 30.1 0.9 100.0
Marital status
Never married 36.1 17.5 14.4 29.9 2.1 100.0 Married or living together 38.9 17.0 16.1 27.6 0.4 100.0 Divorced/separated/widowed 41.2 17.1 13.0 28.4 0.3 100.0 Residence
Urban 38.2 17.0 15.0 29.1 0.7 100.0 Rural 40.0 17.1 15.6 26.8 0.5 100.0
Region
Western 52.4 9.7 5.3 32.3 0.2 100.0 Central 49.4 16.2 12.8 21.6 0.0 100.0 Greater Accra 38.8 17.2 14.7 28.2 1.1 100.0 Volta 28.6 23.8 35.5 10.2 1.9 100.0 Eastern 27.8 18.1 18.3 35.1 0.7 100.0 Ashanti 41.4 12.6 10.6 35.1 0.4 100.0 Brong Ahafo 41.4 16.3 11.1 31.0 0.2 100.0 Northern 49.5 24.8 26.4 28.9 0.4 100.0 Upper East 47.9 27.6 11.4 13.1 0.0 100.0 Upper West 60.6 16.4 5.0 18.0 0.0 100.0 Education level
None 36.7 22.0 17.8 23.1 0.4 100.0 Primary 42.2 18.7 14.2 23.2 1.8 100.0 Middle/JSS/JHS 42.0 15.2 14.7 27.7 0.4 100.0 Secondary 37.4 15.7 14.4 32.1 0.4 100.0 Technical 37.8 14.8 23.5 23.1 0.8 100.0 Higher 31.0 14.8 13.1 40.7 0.4 100.0
Asset quintile
Lowest 35.1 21.6 17.3 25.0 1.1 100.0 Second 40.4 16.2 15.2 28.1 0.2 100.0 Middle 42.7 16.9 14.7 25.4 0.4 100.0 Fourth 39.1 16.0 14.5 29.6 0.8 100.0 Highest 38.1 14.3 14.6 32.3 0.6 100.0
Total 39.1 17.0 15.3 28.0 0.6 100.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
181
Table 59: Views about statement ‘If a woman wears revealing clothes, it is her fault if
she is raped’ (in percentages)
Fully
agree
Somewhat
agree
Somewhat
disagree
Fully
disagree
Do not
know
Total
Sex
Male 56.2 11.0 8.8 23.7 0.4 100.0 Female 65.3 9.5 6.7 18.2 0.3 100.0 Age group (years)
15–19 63.5 9.8 8.5 17.6 0.7 100.0 20–24 64.0 9.2 5.8 20.5 0.5 100.0 25–29 62.9 11.0 6.9 18.5 0.7 100.0 30–39 61.7 11.3 7.8 18.9 0.2 100.0 40–49 59.8 9.2 9.0 21.9 0.1 100.0 50–60 58.8 8.6 7.5 24.9 0.1 100.0 Employment
Self-employed 63.3 9.6 8.3 18.6 0.2 100.0 Employed 54.8 11.7 6.7 26.3 0.6 100.0 Not working 62.3 10.1 6.4 20.7 0.5 100.0 Marital status
Never married 57.2 13.2 9.6 19.2 0.8 100.0 Married or living together 62.2 9.8 7.5 20.3 0.2 100.0 Divorced/separated/widowed 62.1 9.6 6.8 21.0 0.4 100.0
Residence
Urban 57.7 10.1 8.8 23.0 0.5 100.0 Rural 65.8 10.0 6.3 17.7 0.2 100.0 Region
Western 78.6 5.8 1.9 13.5 0.2 100.0 Central 71.3 7.0 6.9 14.8 0.0 100.0 Greater Accra 55.8 10.1 10.7 22.8 0.6 100.0 Volta 45.0 12.1 15.5 26.7 0.7 100.0 Eastern 62.8 10.7 9.3 16.9 0.3 100.0 Ashanti 62.7 10.1 3.7 23.1 0.4 100.0 Brong Ahafo 66.7 8.9 4.9 19.5 0.0 100.0 Northern 48.7 12.9 11.0 27.1 0.3 100.0 Upper East 67.7 15.8 4.1 12.3 0.0 100.0 Upper West 64.1 16.6 7.2 12.1 0.0 100.0 Education level
None 59.5 12.4 8.8 18.9 0.4 100.0 Primary 68.9 9.4 6.1 15.5 0.2 100.0 Middle/JSS/JHS 65.4 8.3 8.1 17.7 0.4 100.0 Secondary 59.6 10.3 6.1 23.7 0.3 100.0 Technical 58.5 7.3 7.7 26.5 0.0 100.0 Higher 43.8 14.0 7.5 34.4 0.3 100.0 Asset quintile
Lowest 59.4 11.7 7.9 20.8 0.2 100.0 Second 64.2 10.1 7.2 18.2 0.4 100.0 Middle 65.6 7.9 7.6 18.4 0.5 100.0 Fourth 61.2 10.0 7.3 21.3 0.3 100.0 Highest 56.7 10.4 8.1 24.4 0.4 100.0
Total 61.6 10.1 7.6 20.4 0.3 100.0
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
182
There was limited discussion about the incidence of sexual violence in the focus group
discussions and individual interviews. Qualitative data on attitudes towards sexual violence
revealed (in the groups that discussed this form of violence), however, that this form of
violence was unambiguously perceived as unacceptable, across all regions. This view is
illustrated, for example, in the following focus group discussion with married men (18–50
years) in the Brong Ahafo region, in the Ahafo Ano North municipal district:
Respondent 3: No, it’s not acceptable. Sex is based on understanding and affection and is only
enjoyable if both partners willingly participate. The woman may not be in the mood so if she’s
not in the mood, how do you enjoy it? Besides how do you beat your partner and have sex with
her the next moment? It’s really cruel.
Respondent 4: No it’s not acceptable. Marriage is based on love and not sex; sex is just one
aspect of marriage so it shouldn’t dictate how you treat your partner. The man should be able to
understand and empathise with the woman in those times that she’s unable for some reason to
give him sexual gratification. There would definitely be other opportunities.
The previous section discussed how patterns of domestic sexual violence differed from other
types of domestic violence. In particular, the results revealed that experiences of sexual
violence were more common among employed, wealthier and more educated individuals
living in urban areas, particularly among men. The research team postulated that this finding
may have emerged because these groups are also more likely to have reported experiences of
sexual violence. The findings in this section go some way towards supporting this. First, there
was very limited discussion about sexual violence in the qualitative data. Second, most survey
respondents showed high levels of acceptance of behaviours defined by sexual violence.
Taken together, these findings suggest that sexual violence may be understood largely as
something that takes place outside domestic relations. Perceptions of what may constitute
sexual violence within domestic relations seem very limited – or hidden. This latter hypothesis
is partially substantiated by the additional finding that larger percentages of female
respondents accept forms of sexual violence, suggesting that women may internalise gender
norms and norms about the acceptance of certain types of behaviour, as discussed in Section
2 and found in other studies.
4.4.4. Attitudes towards domestic psychological violence
As discussed in the sections above on attitudes towards domestic social and physical violence,
the qualitative and quantitative data found that there were circumstances where certain forms
of domestic psychological violence were viewed as defence mechanisms and as a way of
women claiming some limited power in highly unequal domestic partnerships. This finding
was nationally representative in the qualitative data. The following quote, from a participant
in a focus group discussion among young men and women aged 18–25 years in the Eastern
region, captures this finding succinctly:
“Most of the fights are physical but some are verbal. Most of the verbal fights are from the women
and they use them as a defensive mechanism.”
183
Echoing the findings from the qualitative data, data from the GFLHS 2015 revealed that 31 per
cent of acts of insulting, belittling or humiliation of male heads of households (in private or in
front of other people) were perpetrated by their female spouse, partner or former partner. The
remainder were perpetrated by other relations. The incidence of psychological abuse of the
same type against female household heads was lower, around 25 per cent, according to the
GFLHS 2015. The qualitative data provided important explanations for these levels of
incidence. Notably, focus group participants often perceived the exercise of psychological
pressure in the form of verbal abuse as a means of defence and a coping strategy (for women,
in particular) to ward off physical violence, or to claim some autonomy or agency in a context
where women may not have the ability to exert much power beyond the use of words (insults),
their bodies (withholding sex) or their housework (not cooking).
The qualitative data also offered some striking (but not nationally representative) findings that
reveal how men also use psychological violence as a form of retribution. For example, in a
mixed focus group discussion with youth in Kukurantumi (FDG, Eastern region, East Akim
district: male and female group, 18–25 years), a respondent illustrated this finding as follows:
“Let us assume that my wife has offended me but I don’t want to beat her. I can abuse her by not
talking to her for a couple of months.”
In a focus group discussion among young men and women (aged 18–25) in the Ashanti region,
in Obuasi district, a respondent outlined an additional factor that may lead to abuse of women
by men as a form of retaliation:
“Women who are rich or self-made. Who doesn’t respect or appreciate their husbands, especially
if the man is not rich like her, a case in point is when the woman doesn’t appreciate the man’s
gifts, the man feels unappreciated resulting in drunkenness, psychological abuse, etc.”
To explore attitudes towards psychological violence further, the GFLHS 2015 included the
following scenario: ‘A woman repeatedly makes demeaning comments to her husband about
him not making enough money, and not being able to provide for a nice home and gifts for
her. She constantly compares him to his brother, who is a successful businessman.’
Respondents were asked whether this was a scenario that they were familiar with. Nearly half
(48.8 per cent) of respondents stated that this situation was ‘very common’, and 22.9 per cent
that it was ‘somewhat common’. The survey then repeated this scenario with the variation that
the woman was beaten by her husband following the remarks and reported the incident to the
police. Just over half of the respondents (55.8 per cent) agreed that she should have reported
the abuse to the police. Of a total of 1,098 respondents (412 men and 686 women) who thought
the woman should not have reported the incident to the police, 25.2 per cent of men and 32.1
per cent of women mentioned that she should not have reported the beating to the police
because the woman was “the one to blame” for the incident. This finding is in line with the
discussion in the previous sections, where attitudes condoning domestic violence seem more
prevalent among women.
184
4.4.5. Attitudes towards domestic economic violence
Throughout the qualitative data, respondents across all regions expressed strong attitudes
against domestic economic violence. However, the qualitative data drew out interesting
paradoxes. While most forms of domestic violence against women were generally seen as
unacceptable, the data have also shown that men – but not women – were seen to be the main
providers of material well-being to the household and its members. This argument was
forcefully made, in particular, by married women (aged 18–50) in the focus group discussions
that were conducted across all ten regions of the country. The flipside of the argument, which
was made less often by married men, was that the pressure to secure a job and be the sole
breadwinner was also a form of economic violence exerted by women who had not been
restricted from earning an income but who had chosen not to. The following excerpt illustrates
some of the different attitudes women expressed about economic violence. In this particular
conversation, the women articulated one major form of economic violence that was expressed
across the qualitative data – namely, men’s refusal to give women chop money for household
costs:
Facilitator: Is it acceptable if a man refuses to give chop money to his wife?
Respondent 8: No. From the Bible, God said husband should take care of their wives for that
reason any man who marries and doesn’t take care of the woman is not doing right.
Respondent 5: It is not the right thing to do. You have to give her chop money.
Facilitator: Can you explain your answer?
Respondent 2: Because you the man brought the woman to your house so every morning, you
must give her chop money. If you refuse to give me chop money, there will be a fight in the house.
Respondent 3: No. Because he brought me to live with him, so he must take care of me well.
The following extract from a focus group discussion with married women (18–50 years) in the
Brong Ahafo region, in the Ahafo Ano North municipal district, similarly illustrates how
gender norms – and expectations that men should be the ‘breadwinner’ – affect women’s
perception of economic violence. The views expressed by the respondents below were
common across most of the sample and across all regions:
Respondent 3: No, it is not acceptable because it is the man’s responsibility to take care of the
whole family. Since the man cannot starve himself throughout the whole day, he should not also
attempt to starve the rest of the family members. The man cannot say that I cannot refuse giving
chop money to the family. In some instance where the man is not financially sound, he can tell
the woman politely to support him if she has but not to intentionally refuse to give the wife chop
money. It is only a wicked person who will refuse to give the woman chop money.
Respondent 4: It is a big no, because the man is the breadwinner of the family and hence his
responsibility to take care of the family. If for one reason or the other the man does not have
money, he can call the woman to support for that period but not to intentionally refuse to give
her chop money. Considering the way our economy is now, the women are also liable to support
but the responsibility lies on the man to take care of the household.
Respondent 5: It is not acceptable. There are times that the man may genuinely not have money
due to loss of job. Within that period the man can ask the woman to support the house. But, if
the man has money, it not acceptable to refuse to give the woman chop money.
185
Respondent 8: No, it is not acceptable because considering the main responsibilities of a man
at home, providing food is one of their core responsibilities. So if by God’s grace you have money
I don’t see the reason why you should refuse to give chop money. What will happen then is that,
for days that you will refuse to give chop money we will not also leave food for you to come and
eat.
Men felt the strain of these expectations. This excerpt from an interview with an activist in the
Ashanti region describes this view, which was expressed across discussions with men:
“Men tend to experience violence if they are not playing their role well. I am talking about roles
because in Africa, we expect the man to provide. At least make the petty provision, not the
essential. At least to maintain the family. So if you are not maintaining the family, definitely,
the lady turns hostile. When the lady becomes hostile, it rolls unto you. The lady will not allow
you to make love with her, and then she will try to fight you, especially if you are not strong.
And then that is where the men turn to suffer. At times some of the men come here to complain
about the women slapping them, but because of the scriptures and the fact that they are God
fearing, they cannot do anything.”
The denial of chop money also came out very strongly in the quantitative survey. It was the
single most important act of economic violence mentioned against women and men (Tables
23–26): 11.6 per cent of women and 5.7 per cent of men reported having been denied chop
money in the 12 months prior to the survey, despite believing there was enough money
available for those expenses.
4.5. Determinants of domestic violence in Ghana
This section discusses a number of factors the study found to be associated with different
forms of violence against women, men, girls and boys, including where people live, and the
social and economic characteristics of individuals, households and communities. In line with
the sections above, the analysis combined insights from both quantitative and qualitative data.
The sections also include information from multivariate regression analysis, as described in
the methodology section, which has attempted to identify the main determinants of domestic
violence in Ghana in 2015.
To conduct the regression analysis, the research team created a variable for each category of
domestic violence (social, physical, sexual, psychological and economic). This variable takes
the value 1 if the person reported having been exposed to any act of violence within each of
these broad categories in the 12 months prior to the interview; otherwise the variable takes the
value 0. In line with the study’s conceptual framework, the incidence of domestic violence was
regressed on a set of variables reflecting individual, household and community factors, as well
as a set of regional variables.25
25 Each of the ten regions of Ghana is represented by a dummy variable with value 1 if the respondent lives
in the region, and 0 otherwise. The inclusion of regional variables – or regional fixed effects – allowed us to
account for some of the unobserved variation of domestic violence across regions that would otherwise be
186
The variables used in the regressions were derived from the conceptual framework discussed
in Section 2, as follows:
Intergenerational dimension of domestic violence: measured using a binary variable
indicating whether the respondent witnessed each type of domestic violence in their
home during childhood.
Marital status: measured with a variable indicating whether the respondent: (i) was not
married ever; (ii) was in a monogamous relationship; (iii) was in a polygamous
relationship; or (iv) was divorced, separated or widowed.
Socio-economic status: captured by an asset ownership index. The assets included were:
radio, TV, mobile phone, bed, other furniture, cooking utensils, washing machine, fan,
air conditioner, refrigerator, iron, sewing machine, kerosene stove, gas/LPG/biogas
stove, electric stove, open stove, dish/cable TV, wall clock, landline telephone, digital
camera, tablet, computer and mosquito net. The research team chose to measure socio-
economic status through ownership of assets, as this measure is generally recognised
as including less measurement error than income (Deaton, 1997).
Education: measured using a variable that indicated the different levels of schooling
attained by each respondent.
Empowerment: following the existing literature (see review in Justino et al., 2012),
empowerment was measured using data on decision-making. The research team
constructed a decision-making index to identify the involvement of the respondent in
making decisions about the following domains of household life: food purchases, daily
purchases, major purchases and health care (for self).
Patriarchal norms: following the discussion on attitudes towards domestic violence in
the previous sections, the prevalence of patriarchal norms was measured using a
patriarchal gender norm index,26 a standardised index of acceptance of wife-beating27
and an index of tolerance of women’s sexual autonomy.28
included in the error term. Since observations are clustered at enumeration area level, the regressions show
corrected standard errors. 26 This index was constructed based on the number of the following statements for which the respondent
fully or somewhat agrees: ‘A good wife obeys her husband even if she disagrees with him’, ‘Family problems
should only be discussed with people in the family’, ‘To bring up, raise or educate a child properly, the child
sometimes needs to be physically punished’, ‘If a women wears revealing clothes, it is her fault if she is raped’
and ‘It is allowed for men to stare at women they don’t know’. The index was standardised to have a mean
of 0 and a standard deviation of 1. 27 This index correspond to the number of the following cases for which the respondent agrees that a husband
has a good reason to hit his wife: “she does not complete her household work to his satisfaction”“, “she
disobeys him”, “she refuses to have sex with him”, “she asks him whether he has other girlfriends”, “he
suspects she has been unfaithful”, “he finds out she has been unfaithful”, “she contracts a sexually
transmitted disease”, “she goes out without telling him”, “she neglects the children”, “she argues with him”
and “she burns the food”. The index was standardised to have a mean of 0 and a standard deviation of 1. 28 This index corresponds to the number of the following cases for which the respondent agrees that a married
woman can refuse to have sex with her husband: “she does not want to”, “he is drunk”, “she is sick”, “he
mistreats her”, “she is menstruating”, “he does not want to use a condom/contraception” and “she finds out
that he has other girlfriends”. The index is standardised to have a mean of 0 and a standard deviation of 1.
187
Exposure to other forms of violence: measured using an index that included how many of
the following types of violence the respondent had witnessed in their community: theft
or robbery, vandalism, arson, kidnapping or abduction, extortion, bullying, political
violence, highway attacks, harassment, rape, defilement and seduction.29
In addition to the variables above, the pre-analysis of the qualitative and quantitative data
suggested the inclusion in the regression analysis of an additional variable – drinking habits of
the respondents, measured by a binary variable taking the value 1 if the respondent reported
drinking, even occasionally. The regressions control also for the region of residence and
whether the household lives in urban or rural areas.
The regression results are presented in tables F1–F4 in Appendix F, and discussed below, in
conjunction with the findings of the qualitative data.30
4.5.1. Age and intergenerational effects
The literature discussed in Section 2 revealed that experience of domestic violence is often
found to be closely correlated to the age of the victim. In particular, younger women are
generally more likely to report domestic violence than older women (Aizer, 2011; Garcia-
Moreno et al., 2005). The data collected in this study for Ghana are in line with this
international pattern: younger women – and men – were found to be more likely to have
experienced any form of domestic violence than older age groups, as discussed in
subsection 4.1.
The results of the multivariate regressions confirmed these results. Older women were less
likely to report experiencing domestic violence than younger women (Table E1, in
Appendix E). Women aged 50–60 years were 84 per cent less likely to have experienced
domestic violence than women aged 15–19 years (Table E1, in Appendix E). The same
figures for physical, sexual, social and economic violence, respectively, were 99 per cent,
98 per cent, 89 per cent and 73 per cent. Even women aged 20–24 years were significantly
less likely than women aged 15–19 years to experience all forms of domestic violence,
except economic violence, for which there is no statistically significant difference (Table
29 The sample mean for this variable is 0.56, and the median is 0. This indicates that violence tends to
concentrate in a few areas. We, therefore, use a binary variable taking the value 1 if the index of violence
exposure is positive, and 0 otherwise. Similar results in the regression analysis were obtained when we used
the continuous index. 30 The tables in Appendix E display the odds ratios associated with every risk factor, alongside its standard
errors. The odds ratio shows how much more (or less) the odds are of experiencing domestic violence when
a certain risk factor is present, in relation to respondents who did not experience the risk factor. An odds
ratio greater than 1 means that the variable increases the risk of domestic violence. As an example, an odds
ratio of 1.5 for individuals with no education indicates that the odds of experiencing domestic violence are
50 per cent higher among individuals with no education in relation to those with education. An odds ratio
of 2 for individuals with no education indicates that the odds of experiencing domestic violence are 2 times
(100 per cent) higher among individuals with no education than those with education. If instead the odds
ratio is 0.2, this indicates that people with no education are 80 per cent less likely than people with other
levels of education to experience domestic violence.
188
E1, in Appendix E). The odds of men experiencing psychological, physical, social or
economic domestic violence were also reduced as age increased (Table E2, in Appendix E),
though the statistical significance of this association is more variable than the results shown
for women. The odds of men being victims of sexual violence did not vary significantly
with age.
The literature reviewed in Section 2 also discussed how domestic violence is typically
related to strong intergenerational effects (Clark et al., 2010; Gallegos and Gutierrez, 2011;
Bobonis et al., 2013; Pollak, 2004), a result that was also revealed in the Ghana 2008 DHS
study (GSS et al., 2009). This study found similar results for Ghana in 2015. Except for
sexual violence (which may be more hidden from the view of children than other forms of
domestic violence), women who witnessed any form of domestic violence at home as
children were significantly more likely than women who did not witness violence in their
childhood to have been exposed to the same type of domestic violence in the 12 months
prior to the survey (Table E1, in Appendix E). The variations were large: women who
witnessed psychological violence as a child were 2.2 times more likely to be exposed to
psychological violence than women who did not witness this form of violence during their
childhood. The corresponding figures for social and economic violence were almost two
and three times higher. The effect for physical violence was more modest but remained
large: women who witnessed physical violence as a child in their homes were 73 per cent
more likely to be exposed to physical violence than women who did not witness this form
of violence during their childhood.
Men who witnessed social or economic violence in their homes as children were 2.2 times
more likely to have experienced these forms of violence than men who did not witness
these forms of violence in their childhood. Men who witnessed psychological violence as
a child were 80 per cent more likely than men who did not witness psychological violence
in their homes during childhood to have experienced this form of violence as an adult in
the 12 months prior to the survey. The effect was not statistically significant for physical or
sexual violence (Table E2, in Appendix E).
Witnessing violence as a child, therefore, has important generational effects, as these
children were more likely than others to have experienced domestic violence themselves
as adults. The effect also applies to the perpetration of domestic violence. According to the
GFLHS 2015, about 28 per cent of women and men who admitted committing social
violence witnessed social violence as children, compared to 14–15 per cent of those women
and men who did not experience violence as children. Similarly, among men who
witnessed sexual violence as children, almost 30 per cent committed sexual violence over
the 12 months prior to the survey, compared to 2.6 per cent of men who did not experience
this form of domestic violence during their childhood. Across psychological, physical and
economic violence, rates of perpetration are between two and three times higher for
women and men who witnessed domestic violence as children than among individuals
who did not experience domestic violence during their childhood.
189
4.5.2. Socio-economic effects
Section 2 discussed in detail how domestic violence was generally found to be associated
with low levels of income, wealth and education in many parts of the world. These socio-
economic factors were found to be some of the most important determinants of domestic
violence in Ghana in 2015. The vast majority of the research participants identified fights
over money, assets or property as the leading determinants of domestic violence across all
regions. Economic tensions were mentioned as leading to multiple forms of domestic
violence among adults – including between partners, siblings and in-laws. Other factors
reported frequently included education, poverty and the timing of certain economic
events. These findings are discussed in turn below.
Assets, property and inheritance. The data for Ghana discussed in the previous section
showed that women and men in households with lower levels of assets were more likely
than individuals in higher asset quintiles to have experienced physical, psychological and
economic violence. Women in the lower asset quintiles were also less likely to have
experienced social violence than women in higher asset quintiles. However, social violence
among men and sexual violence among both men and women was more concentrated
among those in the top asset quintiles.
The regression analysis in Table E1 (in Appendix E) confirms that the asset ownership
index is negatively and significantly associated with psychological and economic violence
among women: for each of these forms of violence, a one unit increase in the asset index is
associated with a five per cent reduction in the odds of a given woman experiencing
domestic violence. The results in Table E1 indicate that an increase of one standard
deviation of the index is associated with a reduction in psychological and economic
violence among women of almost a third (32.5 per cent) – a very large effect. Asset
ownership is not statistically associated with any other form of domestic violence against
women – indicating that low levels of wealth are only significant risk factors for
psychological and economic violence among women.
The results among the male sample are different. Asset ownership is not statistically
associated with the odds of men experiencing psychological, physical or economic
violence. However, asset ownership is positively correlated with the odds of men
experiencing sexual or social violence (as discussed in subsection 4.1.2): the likelihood of
men experiencing sexual or social violence increased by 3 and 2 per cent, respectively,
when the asset index increases by one unit. This result confirms that men of higher
economic status were more likely than men in lower asset quintiles to have experienced
social and sexual violence in the 12 months prior to the survey.
Even though higher levels of asset ownership lowered the likelihood of some forms of
domestic violence (particularly among women), the qualitative data showed that one of
the major causes of fights in families who experienced domestic violence were disputes
over land assets. This finding was particularly strong in the northern regions, where
190
gender norms played a major role in land conflicts in polygamous contexts that were
arbitrated by customary law. Tensions were particularly high when, for instance, the first
wife of a deceased man had only girls and the second wife had boys: the family of the
second wife could in principle push for the girls not to inherit any of their deceased father’s
property. If properties were not registered, families would have to divide the property
among themselves, which often raised suspicions and mistrust. A female key informant,
who works in litigation in the Northern region, explained when these conflicts occur:
“When one person tries to cheat other family members in sharing family property or when
one person tries to dispose of a family property without the approval and consent of other
family members or to the detriment of the other family members.”
In addition:
“If the man has more than two wives when he dies, the sharing of his property becomes a
problem. Because everybody wants to inherit the man’s property and that brings
disintegration into the family.” (FGD, Upper West region, Jipara district: female group,
18–50 years)
Focus group discussions across all ten regions also referred to fights over assets between
siblings, and within the wider kinship network, when parents died as determinants of
domestic violence. The following excerpt from an interview with an opinion leader in the
Ashanti region further captures some of the dynamics of conflict among siblings that were
frequently cited as causing violence across households in the region:
“Land matter is one of the disagreements and fights in families. The sharing of the land for
farming purposes becomes a problem.”
The same person added:
“There is fighting in the family because of cheating in the sharing of the family properties
like houses, lands, cattle, sheep, goats or trying to take what rightfully belongs to the other
family member(s).”
Education. Section 4.2 discussed how women with primary education were more likely
than women with other levels of education to have experienced physical, psychological or
economic violence. Social and sexual violence was more predominant among women with
secondary education. There was a much less clear pattern of association between education
and domestic violence among men, but men with higher levels education were more at
risk of social and sexual violence.
The multivariate regressions show very varied patterns of association between education
levels and domestic violence among women: (i) women with primary school education are
82 per cent more likely to have experienced physical domestic violence than women with
no education; (ii) women with secondary education are 2.5 times more likely to have
experienced sexual violence than women with no education; (iii) women with primary,
middle and secondary education are, respectively, 82, 89 and 71 per cent more likely to
have experienced social violence than women with no education; and (iv) women with
191
primary education are 50 per cent more likely to have experienced economic violence than
women with no education. There was no statistically significant association between
education levels and the incidence of psychological domestic violence among women
(Table E1, in Appendix E). In general, however, it appears that women with no education
are less likely to experience any form of domestic violence. These results may, however, be
due to the fact that women with more education are also more likely to report having
experienced domestic violence, a factor highlighted in the international literature and in
previous studies in Ghana, as discussed in Section 2.
Education is largely unrelated (or only weakly statistically associated) to the likelihood of
men experiencing domestic violence. The exceptions are economic and sexual violence.
Men with primary, middle, secondary, technical or higher education are, respectively, 2.8,
2.9, 2.9, 3.7 and 2.5 times more likely to experience economic violence than men with no
education. Men with no education seem to be at lower risk of experiencing this form of
violence. Education also matters significantly for the incidence of sexual violence among
men: men with any level of education are substantially more likely to experience sexual
violence than men with no education, for whom the likelihood of experiencing sexual
violence is very close to zero (Table E2, in Appendix E).
Poverty and non-compliance with economic responsibilities. Several respondents in the
focus group discussions and in-depth interviews identified the failure of men to provide
for their family as a major cause of conflict within families, as discussed in the previous
section. Verbal insults and public humiliation by women often followed the failure of men
to provide adequate economic resources, and, in turn, were seen as triggering the use of
physical violence by men against women. This view was widely held across all regions and
most discussion groups, as exemplified in the following statement:
“When a woman nags or complains a lot if her husband is unable to provide her needs, it
obstructs the man from having peace of mind at home. And this makes him go out of the
house often and returns late in the night.” (FGD, Upper East region, Garu-Tenpane
district: female group, 18–50 years)
The GFLHS 2015 data confirmed similar dynamics: 9.4 per cent of men who never denied
chop money suffered from psychological domestic violence; this rate increased to 19.9 per
cent for male respondents who ever denied chop money.
Across all ten regions, the respondents – in particular, older married women who
participated in the focus groups – made links between these forms of domestic violence
and the overall structural economic situation beyond the household, which has left men
(and women) unable to provide for their households. These structural constraints included
lack of education and business skills, as well as the availability of few paid jobs in the wider
economy. In a context characterised by few formal employment opportunities, many men
and women were often compelled to set up their own business, many without having
acquired adequate skills. Frustration and the inability to cope with these wider structural
economic problems were identified as factors that fuelled domestic tensions and resulted
192
in psychological violence committed by women against men, and in physical violence
committed by men against women.
The quantitative data allowed some of these findings to be explored in detail. In particular,
the research team analysed the association between employment status and domestic
violence, because employment opportunities are typically a good indicator of the overall
wealth of the economy, and a determinant of households’ economic welfare. We discussed
in section 4.2 how, overall, women and men who were not working at the time of the
survey were more likely to experience any type of domestic violence than women and men
who were employed or self-employed. The exception was sexual violence – as discussed
before, this form of violence was more prominent among women and men who were
employed.
However, the regression analysis in Tables E1 and E2 (both in Appendix E) did not support
these results. These tables show that there is almost no correlation between the
employment status of women or men and their experiences of domestic violence,
suggesting that other factors associated with employment status may be driving the results
(for instance, asset ownership or education). The main exception is domestic economic
violence: (i) women who were not working were 43 per cent more likely to have
experienced economic violence than self-employed women; and (ii) men who were
employed or not working were, respectively, 2 times and 2.5 times more likely to have
experienced economic violence than self-employed men. In line with previous discussions,
employed men are at greater risk of sexual violence (Table E1, in Appendix E).
Timing of economic events and expenditures. Related to the findings above, domestic
violence was also found to be linked to specific moments, festivals, events and seasons,
such as Christmas, the beginning of the school year and pay day. An opinion leader in the
Eastern region summarised this issue, widely reported across all regions:
“Yes, normally some of these fights happen during funeral times, festivals and also when
schools resume and parents have to pay school fees, sew new school uniform and other
stationery.”
In addition, conflicts about expenses at the start of school were reported across Ghana.
Respondents from all regions mentioned that women usually buy the teaching and
learning materials for their children with financial support from their husband. When
children are not properly clothed or are missing items, they might be sent home, bringing
shame to their family. Fighting about school expenses predictably takes place, as illustrated
in this quote, which reflects nationally representative findings:
“When schools reopen, usually within the second and third weeks. It is during this period
that we need the children’s learning materials. So this information is put on paper to be given
to their parents, and this can cause fights if fathers refuse to follow that.” (IDI, opinion
leader, Northern region, Sagnerigu district)
193
In farming communities, harvesting or planting seasons were often mentioned as a cause
of conflicts between couples, particularly when views diverged as to whether to send
children to school or keep them at home to help on the farm. The following respondent
expresses a view often encountered in rural areas:
“Mostly in the rural and farming communities, it happens between the months of May to
October, when the first rains fall and the farmers plant their crops. Often the man may ask
the children to stay home from school and help with the farm work, the woman may object to
this and it often results in violence.” (IDI, opinion leader, Northern region, Sagnerigu
district)
Spending monthly salary payments on personal pleasure – especially on drinking – was
found to lead to domestic conflicts. A community leader articulated this nationally
representative finding, saying:
“When workers are paid at the end of the month and the man does not cater for the family,
but is seen at drinking spots, it can also cause fights within the families.” (IDI, Eastern
region, East Akim district)
Alcoholism was mentioned in every region, and almost every district, as a particularly
problematic determinant of domestic violence, with men who consume excessive levels of
alcohol being more likely to commit violence against their wives. This was more frequently
articulated to be a problem in communities where pito or akpeteshi (local beers/spirits)
consumption was high in the poorer northern regions. For example, a community leader
explained that:
“Some of the men force the women to go and work on farm, whiles they sit at home,
sometimes drinking ‘pito’ – that is the local alcohol drink.” (IDI, Upper East region, Garu
Tempane district)
The regression analysis in Table E1 (Appendix E) showed that alcohol consumption was
significantly associated with 39 per cent higher odds of women experiencing economic
violence. Alcohol habits are associated with a 66 per cent increase and a doubling of the
odds of men experiencing psychological and sexual violence, respectively (Table E2, in
Appendix E). These results do not indicate a causal relationship between alcohol
consumption and domestic violence but strongly suggest that the risks of domestic
violence incidence increases when alcohol consumption is high.
4.5.3. Marital status
Several international and Ghana-based studies reviewed in Section 2 have shown that
marital status is an important factor associated with domestic violence. The discussion in
Section 4.2 on incidence of domestic violence showed mixed patterns: married women (or
those living with a partner) were at higher risk of sexual or economic domestic violence
than unmarried women, whereas divorced, separated or widowed men were more at risk
of social or sexual violence than married or never-married men. All other forms of violence
were more prominent among non-married women and men.
194
These results change substantially in the regression analysis in Tables E1 and E2 (Appendix
E). The results show that, in general, not being married reduces the likelihood of having
experienced domestic violence among women and men. Furthermore:
Divorced, separated or widowed women were twice as likely to have experienced
physical violence than women who were never married.
Married women or divorced, separated or widowed women were, respectively, 1.8
times and 2 times more likely to have experienced economic violence than non-
married women.
Married or divorced, separated or widowed men were, respectively, 6 times and
almost 8 times more likely to have experienced sexual violence than non-married
men, and almost 3 times and 1.7 times more likely to have experienced social
violence than men who were never married.
Table E1 (Appendix E) shows that women in polygamous relationships were almost twice
as likely to experience psychological violence than women who were never married. These
women were also 3.5 times more likely to have experienced physical violence and 4.2 times
more likely to have experienced economic violence. Among men, polygamy increases
considerably the odds of sexual or social violence against them (Table E2, in Appendix E).
However, the association between the respondent’s marital status and experiences of
different types of domestic violence could reflect reverse causality, as divorce and
separation, for instance, may result from domestic violence. Disentangling these issues
through quantitative data alone is very challenging. The qualitative data (and some
additional quantitative data) suggest some plausible mechanisms that may explain these
correlations, including access to sex within marital relations, extra-marital affairs and
forced marriage, as discussed in more detail below.
Access to sex within marital relations. The work of Ellsberg et al. (2001), cited in Section
2, showed how, in many parts of the world, marriage is seen as guaranteeing men
unconditional sexual access to their wives. Conflict over sex was a ubiquitous finding in
the qualitative data, prevalent across all regions, and in both rural and urban areas. The
different compositions of the focus groups enabled frank discussions about these sex-
related conflicts, particularly in the single-sex groups of married women and of married
men (aged 18–50, and aged 50+). The principle source of conflict, across all regions, centred
on men’s desire for sex at times when they were unable, or unwilling, to provide for the
members of the household, including giving their wives chop money. This nationally
representative finding was highlighted as one of the major causes of violence – particularly
psychological violence and controlling behaviour.
This finding suggests a close link between economic and marital determinants of domestic
violence: women described how they expressed anger against their partner by refusing sex,
while men responded or initiated marital conflicts by refusing to fulfil their economic
obligations. In particular, most respondents across all regions agreed that both women and
195
men use sex as a weapon that has many forms, including being forced to have sex or being
denied sex. Sex is part of an intricate power play between men and women, as explained
in this interview with an education professional:
“Fear can also make women feel [that] they need to accept the dominant or violent ways of
the men. She is afraid to be beaten, afraid to be denied food and afraid of being driven out of
the matrimonial home, and some even fear being denied sex, especially in the polygamous
marriage. The man can proudly say ‘I will not have sex with you.’” (IDI, Northern region,
Sagnarigu district)
The quantitative survey included a number of questions about this important issue.
Opinions diverged widely about whether women have a right to refuse sex. For instance,
8.2 per cent of respondents (9.8 per cent of women and 5.7 per cent of men) felt that the
refusal of sex was a justifiable reason for wife-beating (Table 55). Employed respondents
(4.6 per cent) were less likely to agree that the denial of sex was a reason for wife-beating
than self-employed (9.2 per cent) and non-working respondents (8.1 per cent); as were
respondents living in urban areas (4.8 per cent, compared to 11.7 per cent in rural areas),
in the Greater Accra region (1.7 per cent, compared to 24.4 per cent in the Northern region,
24.0 per cent in the Upper East region and 29.4 per cent in the Upper West region), with
higher levels of education (2.5 per cent, compared to 17.5 per cent among those with no
education) and in the highest asset quintile (3.1 per cent, compared to 13.4 per cent in the
lowest quintile).
Extra-marital affairs. Men and women who engaged in extra-marital affairs were
perceived to be at risk of domestic violence. The qualitative data indicated that this
violence differed by sex, but that it was ubiquitous across all regions. Women who had
extra-marital affairs were more likely to experience physical violence perpetrated by their
partners, whereas men were more likely to experience psychological violence and
controlling behaviour from their partners. There were, however, some striking anomalies
in the qualitative data, which suggested that this gendered dichotomy of experiences of
violence might not always hold true. In a focus group with young men and women (aged
18–25) in the Greater Accra region, this young person recalled a case of a man who found
out that his wife was cheating:
“What happened was that the man was called by the friend that he saw the wife somewhere
with another man. Instead of the man asking the woman where she went, upon her return,
he pounced on her and started beating her and insulted her. The woman also went into the
room and got herself an iron and hit the man with it.” (Mixed FGD, Gbawe,
unmarried/youth (18–25 years)
The findings from the survey differed somewhat from the qualitative findings but are
equally paradoxical: 0.4 per cent of the survey respondents agreed that unfaithfulness
would be an acceptable reason to hit women, but 7.5 per cent mentioned that the suspicion
of unfaithfulness was an acceptable reason for wife-beating (Table 55). This finding
requires further future research, as the data collected are not sufficient to fully understand
these findings: unfaithfulness emerges as a risk factor for domestic violence, but
196
perceptions and responses vary substantially across types and perpetrators of domestic
violence.
Forced marriage. Forced marriage may itself be considered a form of domestic violence
perpetrated by families against girls with the consent of their parents. While not widely
reported in the qualitative data, respondents in the Northern region reported that parents
would be more likely to arrange an early or forced marriage when girls were found to be
underperforming in school and paying attention to boys. While non-representative, this
narrative was iterated by a number of community leaders and activists in interviews. For
example, according to a faith leader in the Northern region:
“As we sit now, there is a girl whose parents have arranged with an old man who is already
married to two wives to support their daughter in school and marry her. The girl is now in
senior high (SHS) and this is creating a serious fight between the girl and her parents.” (IDI,
Northern region, Sagnarigu district)
Similarly, an activist working on education explained that early marriages are a form of
violence that reflect structural poverty and absent fathers:
“Some of the children who are female may look for support from outside the house from men.
No man will give out money free. Therefore, these girls may end up being impregnated. This
is where early marriage comes in. She may be forced to be given out to marriage at that tender
age, if she attends school, she may drop out of school making her situation worse and her
future bleak.” (IDI, Northern region, Sagnerigu district)
This issue was only raised in the discussion groups in the northern regions of Ghana.
4.5.4. Geographical effects
The results showed several differences in how domestic violence and attitudes towards
domestic violence were reported across the different regions of Ghana. The discussion in
Section 4.2 highlighted that women and men living in urban areas were at greater risk of
experiencing social or sexual violence, whereas women and men living in rural areas were
at greater risk of experiencing physical, psychological or economic domestic violence. The
experiences of women and men across the ten regions were very different, and no
immediate pattern stood out.
The regression analysis in Tables E1 and E2 (Appendix E), which control for a range of risk
factors, showed no statistically significant difference in the likelihood of domestic violence
experienced by women in urban and rural areas (the coefficients were not statistically
significant). The only exceptions are that: (i) women in rural areas were 39 per cent more
likely to suffer from economic violence than women in urban areas; and (ii) men in rural
areas were 2.2 times more likely to have experienced physical violence than men living in
urban areas. In all other cases, the effect of living in a rural or in an urban location was not
found to be associated with domestic violence.
197
Most of the regional differences observed in women’s experiences of physical and social
violence disappear in the multivariate analysis (Tables E1 and E2 in Appendix E). The only
associations found between regions and the incidence of domestic violence included: (i)
psychological violence against women in the Central, Eastern, Ashanti and Upper regions,
and against men in the Eastern, Brong Ahafo and Upper East regions; (ii) physical violence
against women in the Eastern and Upper East regions; (iii) sexual violence against women
in the Central and Greater Accra regions; (iv) social violence against women in the Eastern
region and against men in the Brong Ahafo, Northern and Upper East regions; and (v)
economic violence against women in the Central, Greater Accra, Volta, Ashanti, Northern,
Upper East and Upper West regions, and against men in the Volta and Upper East regions.
Table 60 summarises this information.
Table 60: Regions significantly associated with increased experience of domestic
violence (by type of domestic violence and sex)
Social
violence
Physical
violence
Sexual
violence
Psychological
violence
Economic
violence
Women Eastern Eastern
Upper East
Central
Greater
Accra
Central
Eastern
Ashanti
Upper East
Upper West
Central
Greater Accra
Volta
Ashanti
Northern
Upper East
Upper West
Men Brong Ahafo
Northern
Upper East
Eastern
Brong Ahafo
Upper East
Volta
Upper East
Source: Ghana Family Life and Health Survey (GFLHS) 2015 and DHS 2008.
Note: Based on results reported in Tables E1 and E2 (Appendix E).
However, for both the male and female samples, regional effects were not collectively
statistically significant (Tables E1 and E2, in Appendix E).31 These results suggest that, once
the regressions control for a range of variables, there is no regional exceptionalism in
patterns of domestic violence. In other words, the differences in domestic violence
incidence discussed in Section 4.2 are due to a range of factors – possibly levels of poverty
or economic development – but not to intrinsic geographical characteristics.
4.5.5. Patriarchal and gender norms
Section 2 discussed in detail how patriarchal norms and perceptions of masculinity that
support the persistence of certain gender norms may be closely associated with experiences
of domestic violence. The association of domestic violence in Ghana with patriarchal
31 With the exception of economic violence, in the regression for men, for which regional effects are
marginally significant.
198
cultural norms was evident in the qualitative and quantitative data collected across all
regions and cut across the various socio-economic backgrounds of the respondents.
The qualitative and quantitative data found that social norms concerning appropriate
behaviour for women were a significant trigger for domestic violence by men against
women in all regions. In the qualitative interviews and focus groups, the research team
asked which expectations of masculinity make men behave in certain dominant or violent
ways. The male and female respondents, in every region, discussed in detail which societal
gender norms and expectations shaped appropriate and inappropriate behaviour for both
men and women. The dominant view was that domestic violence was associated with non-
compliance with these gender norms. As discussed in the section above on attitudes to
violence, the quantitative and qualitative data found that different forms of domestic
violence were used to ‘punish’ men and women, and young people, when they had been
found to contravene normative expectations of appropriate behaviour. For instance, the
qualitative data found that men were more likely to become violent when their expectation
of being obeyed was not fulfilled by their wives, or when wives failed to perform expected
domestic roles, such as ensuring their meals were ready on time, or refused to have sex. A
community elder in the Ashanti region explained a widely held view across all regions:
“Men, like children, expect to be pampered at home. Thus, if they do not get this from their
wives, it makes them act violently in the house. Also when they are refused sex it makes them
go wild in the house. The other main reason why men act violently at home is when women
do not [fulfil] their wifely duties. For instance, you know the man sometimes come back from
town in the company of friends and when the wife always leave the house untidy, he will
fight with you.”
The respondent added:
“The man expects that he would be accord[ed] the needed respect and dignity at home by
both his wife and children so, if he is not getting it, it becomes a problem. It also happens
when the man sees that the woman is inciting the children or trying to organise the children
against him. The man also expects that his instructions must be obeyed at home.”
The qualitative data from all regions, and especially from the Northern region, also found
that societal norms governing expectations about women made them more accepting of
abusive relationships. Both male and female respondents pointed to the ‘expected’
submissiveness of women within conjugal relationships, women’s roles as primary
caregivers (for children) and their economic dependence on men as reasons for them to
accept violence. For example, a health practitioner in the Ashanti region discussed these
gender dynamics in an interview, saying:
“At times they can say I am the head of the house. Whatever I instruct you to do, you do it.
I give you money daily for you to eat. I clothe you. That is why at times they do that. It could
also be our bringing up as Ghanaians and also what they tell married couples on the day of
their marriage. Whatever your husband tells you to do, do it. Even if your husband beats
you, stay in the marriage. The marriage counselling she received says that the husband is the
head. Even in our churches they say the husband is the head. Maybe it could be that the
199
woman does not wash the man’s clothes, or not giving the man food to eat. When you also
deny the man sex, he can behave in this violent way.”
This view emerged across the focus group discussions in all ten regions, and is illustrated
by the following exchange in the Central region:
Respondent 8: We are told from the word go that we have to be submissive so we seem to
have no other choice. Sometimes is because of the children so we don’t even care what happens
to us. I will plead with men to have patience for women.
Respondent 7: Sometimes is because the man takes care of all finances so we can only be
submissive no matter what the case may be but in cases where at least the woman is engaged
in some occupation and can also support financially then the domination will be minimised.
Respondent 2: The men take care of us, so we cannot even go anywhere without their
permission.
(FGD, Central region, Anomabo district: male and female group, 18–25 years)
The quantitative data allowed the research team to analyse the effect of gender norms on
domestic violence using four distinct variables: a patriarchal gender norms index; an index
that measured tolerance towards wife-beating; an index that measured acceptance of
women’s sexual autonomy; and a measure of decision-making power within the
household.
The results in Table E1 (Appendix E) show that women who exhibited higher levels of
acceptance of wife-beating were 14 per cent more likely to have experienced economic
violence than women with lower acceptance of wife-beating. As mentioned before, this is
not a causal effect and may well suggest (as discussed in Section 4.2 on attitudes) that
victims of violence internalise and accept gender norms that perpetuate or justify domestic
violence. Women who expressed a higher acceptance of women’s sexual autonomy were
16 per cent more likely to have experienced psychological and social violence, 17 per cent
more likely to have experienced physical violence, and 11 per cent more likely to have
experienced economic violence than women with a lower acceptance of women’s sexual
autonomy. This latter finding suggests, as discussed in the qualitative data analysis above,
that women who are more likely to challenge accepted gender norms may be at risk of
domestic violence because they may be punished for their non-compliance with
established norms of behaviour. The results in Table E1 (Appendix E) show no statistically
significant association between domestic violence against women and the patriarchal
gender norms index or the levels of decision-making within the household.
The results in Table E2 (Appendix E) show that men who expressed adopting stronger
patriarchal gender norms were 26 per cent less likely to report social violence than men
who did not accept such norms. However, men who displayed higher levels of tolerance
towards wife-beating were 21, 47 and 30 per cent more likely to have experienced
psychological, physical or economic violence, respectively, than men who reported lower
levels of tolerance towards wife-beating. This result may potentially reflect retaliation by
women who are beaten by their husbands, as discussed in the previous section. Men who
200
exercised more decision-making power within their households were 34 per cent less likely
to experience social violence than men who exercised less levels of decision-making within
the household.
Taken together, and in conjunction with the qualitative data discussed above, these results
suggest that gender norms shape the extent of domestic violence in Ghana. This finding is
in line with recent research by Heise and Manji (2016), who have shown how social norms
play important, albeit different roles, for actors functioning at different levels of the social
ecological model.
4.5.6. Exposure to other forms of violence
Some recent evidence from conflict-affected countries has suggested that domestic
violence may be associated with other forms of violence that take place across countries
and within communities. The survey questionnaire included questions on various forms
of violence that may occur in the Ghanaian society, which were aggregated using an index
that included how many of the following types of violence the respondent had witnessed
in their community: theft or robbery, vandalism, arson, kidnapping or abduction,
extortion, bullying, political violence, highway attacks, harassment, rape, defilement and
seduction. The regressions in Tables E1 and E2 (Appendix E) show that exposure to
violence in the community was significantly associated with higher levels of domestic
violence. The odds of victimisation among women increased by 77 per cent for
psychological violence, 49 per cent for physical violence and 57 per cent for economic
violence when the respondent lived in a community that experienced at least one form of
violence. The effect was stronger for sexual violence and social violence, for which the odds
of having experienced this violence were 3.6 times and 1.9 times higher, respectively, for
women exposed to violence in their communities than for women in communities that did
not experience other forms of violence.
The results showed an even stronger association between violence in the community and
incidence of domestic violence among men. Men who reported some violence in their
communities were between two and three times more likely than men living in more
peaceful communities to report having experienced any form of domestic violence.
The qualitative data emphasised, in addition, that in the northern regions of Ghana
chieftaincy disputes between extended royal family members can erupt into community-
wide violence if, for instance, the members of the extended family or even siblings fight
over the right to the throne, or disputes and fights erupt between other family members
who may support different candidates:
“Sometimes too if the man and woman’s family are fighting over the right to the throne, it
often results in fights and conflicts between them and sometimes a break of the union by the
two families even when they love each other.” (IDI, Northern region, Sagnerigu district,
Gbolo opinion leader)
201
Chieftaincy disputes were only mentioned in the northern regions, where they are
described as “a major cause of disagreements and fights in this community.” (FGD, Upper East
region, Garu-Tempane district: male group, 18–50 years)
Despite the strong association between domestic violence and other forms of community
violence, the study is not able to indicate the direction of causation. On the one hand,
communities that experience high levels of community violence may also experience high
levels of domestic violence because violence becomes an accepted way of behaviour. On
the other hand, it is possible that experiencing violence within the family and immediate
social relations may predispose individuals to other forms of violence. This result merits
further research in future analyses of the GFLHS 2015.
4.6. Consequences of domestic violence in Ghana
This section addresses the research question about the consequences of domestic violence
in Ghana. The main consequences of domestic violence emphasised in the literature
reviewed in Section 2 included effects on physical health, effects on mental health and
emotional well-being, effects on concentration during and take-up of day-to-day tasks, and
indirect effects on children. The qualitative and quantitative data collected in Ghana for
this study revealed similar patterns, with a strong emphasis on the negative physical and
mental health impacts of domestic violence. As with other sections, this analysis draws on
complementarities between these different data. Multivariate regression analyses were
also undertaken to explore the physical and mental health consequences of domestic
violence. The study used three self-reported measures of physical health: health status,
illness and chronic condition. Health status was captured by the following question: ‘In
general, would you describe your general health as very good, good, fair, poor or very
poor?’ ‘Very good’ and ‘good’ categories were grouped together into a binary variable
indicating ‘good health’. Illness was captured by the following question: ‘In the last 30 days,
have you felt sick or unwell?’ Chronic condition was measured by the following question:
‘Do you have any chronic condition?’.
The study also used the K6 scale of mental illness developed by Kessler et al. (2003) to
assess mental health. The K6 scale is constructed from a survey module that asked how often
in the past 30 days the respondent felt any of these six symptoms: nervous, hopeless,
restless, so depressed that nothing could cheer them up, that everything was an effort, and
worthless. For each of these six questions (K6), possible answers included: ‘all the time’,
‘most of time’, ‘some of the time’, ‘a little of the time’ and ‘none of the time’. A respondent
who answered ‘all the time’ across the six items was attributed a maximum K6 score of 24
(6*4). Respondents with K6 scores exceeding 13 were considered to be suffering from
serious mental illness (SMI) (Kessler et al., 2010). The regression analysis used both the
continuous (K6 scale) and the binary (SMI) variables in the analysis. Regressions, in
addition, control for age, education, marital status, socio-economic status, residence and
violence in the community, as well as food security (which takes the value 1 if household
members went to bed hungry, even if only rarely, during the last lean season), source of
202
drinking water (distinguishing between piped water into the dwelling; public taps, wells
and springs; surface water; and bottle/sachet water) and access to toilet (distinguishing
between private toilet; shared toilet between a few families; public toilet; and no toilet).
4.6.1. Physical health effects
To measure the impact of domestic violence on respondents’ health, the quantitative
survey asked respondents who reported having experienced any act of sexual or physical
violence whether the act had resulted in injuries. A total of 182 men and 315 women across
the whole GFLHS 2015 experienced an act of sexual violence in the 12 months preceding
the survey, of which 32 incidents were domestic acts of violence against men, and 78
against women. Out of these, nine women and one man reported having received injuries.
Seven women were injured once or twice, and the man and two women more than five
times. Four of the respondents needed health care following their injuries; one of the
women needed health care three times in the 12 months prior to the survey.
Fifty men and 192 women survey respondents reported having experienced physical
domestic violence in the 12 months that preceded the survey. Injuries were reported by 114
of these respondents: 14 men and 77 women were injured once or twice during that time
frame, one man and 16 women were injured between three and five times, and one man
and five women were injured more than five times. Forty-six women and ten men were
hurt badly enough to have needed health care: 32 once, and 24 more often.
Other studies conducted in Ghana (e.g. Pool et al., 2014) have shown that violence against
women is associated with serious health consequences in terms of pregnancy outcomes.
The GFLHS 2015 found that 30 of the 192 women who reported having been physically
assaulted were hit while pregnant (15.6 per cent), resulting in a miscarriage rate of 10 per
cent (three women) and neurological complications to the foetus in 6.7 per cent (two
women) of the cases. Another 6.7 per cent reported other, non-specified, effects on the child
or mother.
Domestic violence also appears to have exerted an adverse impact on general health
outcomes: 43.8 per cent of women who had been physically assaulted declared having been
ill in the 30 days prior to the survey – a rate that was one third higher than among women
who were not physically assaulted (31.2 per cent). Also, 42.3 per cent of women who had
experienced psychological violence reported ill health, compared to 30.9 per cent of
women who did not experience this form of domestic violence.
The social ecological model adopted by this study understands the consequences of
domestic violence within the complex interplay of individual, relationship, social, cultural
and environmental factors. This understanding was evidenced in all focus group
discussions and in-depth interviews that addressed the consequences of domestic violence,
and particularly emphasised in the younger (mixed unmarried, 18–25 years) and older
groups (mixed, 50+ years). These two groups mentioned, in addition, that the health
203
consequences of domestic violence were felt not just by the individuals involved (victim or
perpetrator) but also by witnesses of the violence within the household, family and
community.
At the individual level, domestic violence was described as affecting several health
complications such as increased blood pressure, mental disorders, depression, eating
disorders, sleeplessness, alcoholism and suicidal thoughts. These responses sum up the
range of views on the health consequences of domestic violence:
“Psychological violence is also likely to cause the victim stress and pressure because
continuously he or she would have to be thinking and feel isolated. These could also lead to
several health complications such as increase in blood pressure, mental disorders, depression,
eating disorders, sleeplessness, alcoholism, suicidal thoughts and others. Sexual violence also
results in unwanted pregnancy, chronic pelvic and abdominal pain and contraction of STIs.”
(IDI, opinion leader, Kukurantumi/Koforidua East Akim district, Eastern region)
Physical violence was commonly noted as affecting the physical health of women and men.
For men in particular, and in all regions, alcoholism was consistently mentioned by all
categories of focus group discussions and in-depth interviews as both a consequence and
a cause of violence (as discussed also in the previous section). This nationally
representative view is well expressed by a respondent in a focus group discussion in the
Greater Accra region:
“It [domestic violence] affects the man and woman mentally, emotionally, psychologically,
to the extent that the men start drinking alcohol and the woman suffer from depression.”
(FGD, mixed 50+, Tomefa, Ga South district, Greater Accra region)
Focus group discussions across all regions mentioned that household and community
members who witnessed these outbursts of violence were also negatively affected by the
violence, as illustrated in these two poignant quotes:
“[The] aged in such environments could have their blood pressures rising upon hearing of a
fight.” (FGD, mixed 50+, Gbawe, Ga South district, Greater Accra region)
“[Domestic violence] causes emotional distress to neighbours or family members who may
be in the household.” (FGD, married/living together women 18–50, Gbolo, Sagnerigu
district, Northern region)
These responses reveal the relationship between individual and contextual factors and
demonstrate how domestic violence affects multiple social spheres.
Table 61 shows the regression results for the effect of each form of domestic violence –
psychological, physical, sexual, social and economic – on the physical and mental health
of women. The results show that:
Women who experienced psychological violence were 29 per cent less likely to
assess their health as very good or good than women not affected by violence.
Women who experienced physical violence were 67 per cent more likely to have
had an illness in the month prior to the survey than women not affected by violence.
204
Women who experienced social violence were 32 per cent less likely to have
assessed their health as very good or good than women not affected by violence.
Women who experienced economic violence were 40 per cent less likely to assess
their health as very good or good, 36 per cent more likely to report having been ill
in the month prior to the survey and 66 per cent more likely to have reported
suffering from at least one chronic condition than women not affected by violence.
Table 61: Relationship between experience of domestic violence and physical health
(women)
(1) (2) (3)
Health status
(1=very
good/good)
Illness (1=ill
during last
month)
Chronic conditions
(1=yes)
Logit Logit Logit
Psychological violence 0.71* 1.19 1.10
(0.13) (0.19) (0.23)
Physical violence 0.71 1.67*** 1.38
(0.16) (0.29) (0.34)
Sexual violence 0.97 1.30 0.80
(0.34) (0.39) (0.29)
Social violence 0.68** 1.41** 1.20
(0.13) (0.20) (0.21)
Economic violence 0.60*** 1.36** 1.66***
(0.091) (0.19) (0.30)
Observations 2,971 2,971 2,963
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Note: The table shows the odds ratios. Standard errors in parentheses are clustered within enumeration areas
(EA). * p < 0.1, ** p < 0.05, *** p < 0.01. Each column represents the results of a regression equation that includes
all controls included in the tables in Appendix E.
The results in Table 61 show that sexual violence among women did not have any
statistically significant effects on their health. In stark contrast with the results for women,
Table 62 shows that none of the forms of domestic violence were statistically related to the
physical health status of men.
205
Table 62: Relationship between experience of domestic violence and physical health
(men)
(1) (2) (3)
Health status
(1=very
good/good)
Illness (1=ill
during last
month)
Chronic conditions
(1=yes)
Psychological violence 0.54 0.76 0.94
(0.26) (0.29) (0.46)
Physical violence 0.67 1.55 0.99
(0.71) (0.80) (0.69)
Sexual violence 0.57 2.23
(0.48) (1.67)
Social violence 0.43 1.00 1.63
(0.26) (0.33) (0.78)
Economic violence 0.55 1.37 1.32
(0.25) (0.41) (0.64)
Observations 695 695 686
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Note: The table shows the odds ratios. Standard errors in parentheses are clustered within enumeration areas
(EA). * p < 0.1, ** p < 0.05, *** p < 0.01. Each column represents the results of a regression equation that includes
all controls included in the tables in Appendix E.
The higher incidence of health effects of domestic violence among women may be due to
women being more vulnerable to physical assault than men. As a respondent poignantly
pointed out:
“The abusers are natural[ly] stronger. So if it comes to the physical attack very few women
will be able to outwit men. Most times the men overpower the women.” (IDI, opinion
leader, Jirrapa district, urban Jirrapa)
As discussed in Section 4.1, this view was common across all regions, as illustrated by this
quote:
“There is a difference in the violence that men and women experience because a man is
stronger than a woman. As a result, women are mostly hurt in violence than men.” (FGD,
Upper West Jirrapa district, rural Wuling)
And, when
“…the violence is physical, it can cause injury to the victim which could lead to permanent
deformity in the long term.” (FGD, married/living together men, Gbolo: Sagnerigu
district, Northern region)
206
The qualitative data showed that the consequences of physical abuse, including slapping,
pushing, kicking, strangling and using weapons, varied from minor injuries to death.
Physical injuries often led to hospitalisation and incapacitated victims:
“Women experience a lot of domestic violent cases more than men. A lot of women end up
losing their lives because they are most often at the receiving end of these brutish attacks.”
(IDI, opinion leader, Goaso, Ahafo North district, Brong Ahafo region)
The majority of respondents interviewed who work in health services mentioned having
seen permanent disability, scars and even death as a result of physical violence,
particularly against women.
4.6.2. Mental health and emotional effects
The survey data reported a very strong correlation between exposure to domestic violence
and mental health. Using the K6 scale for depression and anxiety, Table 63 shows that the
likelihood of suffering from serious mental illness (i.e. having K6 scores exceeding 13) was
considerably higher for women and men exposed to domestic violence in the 12 months
prior to the survey than for those who were not. Women who reported serious mental
illness were 26.7, 24.2, 21.9, 19.3 and 26.3 per cent more likely to have experienced,
respectively, psychological, physical, sexual, social or economic violence than women who
did not report serious mental illness. The association between serious mental illness and
domestic violence was generally lower among men, but still significant across all
categories, with the exception of sexual violence.
Table 63: Percentage of respondents who have experienced domestic violence
reporting serious mental illness (by type of domestic violence and sex)
Psychological
violence
Physical
violence
Sexual
violence
Social
violence
Economic
violence
Women
Yes 26.7*** 24.2*** 21.9*** 19.3*** 26.3***
No 11.0 11.7 12.3 11.6 10.6
Men
Yes 11.7* 19.0*** 6.9 12.7** 15.7***
No 7.9 7.9 8.2 7.8 7.6
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Note: Pearson Chi-squared test of equality of means between ‘yes’ and ‘no’ categories: *** p<0.01; ** p<0.05;
*p<0.1
Table 64 illustrates further, using regression analysis, the consequences of domestic
violence on mental health among women. Column 3 shows the effect of each form of
domestic violence on their K6 score (women exceeding a K6 score of 13 were considered
to be suffering from serious mental illness). Column 4 shows the same results using a
207
binary variable (which takes the value 1 if the woman was considered to be suffering from
serious mental illness, and 0 otherwise). The results were as follows:
Women who experienced psychological violence had a K6 score that was 5.6
percentage points higher than women who did not experience psychological
violence, and were twice as likely to suffer from serious mental illness.
Women who experienced social violence had a K6 score that was 2.4 percentage
points higher than women who did not experience social violence, and were almost
1.5 times more likely to suffer from serious mental illness.
Women who experienced economic violence had a K6 score that was almost 10
percentage points higher than women who did not experience economic violence,
and were 94 per cent more likely to suffer from serious mental illness.
Table 64: Effect of domestic violence on men and women’s mental health
(1) (2) (3) (4)
Men: K6
score
Men: Serious
mental illness
(K6 > 13)
Women: K6
score
Women:
Serious
mental illness
(K6 > 13)
OLS Logit OLS Logit
Psychological violence 1.22 0.35 5.63*** 2.07***
(0.73) (0.23) (2.58) (0.43)
Physical violence 4.12 5.86*** 2.07 1.18
(5.39) (3.75) (1.10) (0.27)
Sexual violence 0.77 n.a. 2.95 0.94
(1.26) (2.45) (0.38)
Social violence 1.04 0.80 2.41*** 1.47**
(0.75) (0.50) (0.81) (0.24)
Economic violence 3.05 1.64 9.95*** 1.94***
(2.09) (0.90) (4.44) (0.34)
Controls YES YES YES YES
Observations 704 687 2,972 2,972
R-squared 0.12 0.16
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Note: The table shows the odds ratios. Standard errors in parentheses are clustered within enumeration areas
(EA). * p < 0.1; ** p < 0.05; *** p < 0.01.
The results show that physical and sexual violence had no statistically significant effects
on the mental health status of women. In addition, Columns 1 and 2 of Table 64 show that
(as with physical health) domestic violence did not have statistically significant effects on
the mental health status of men. The only exception was for physical violence, where those
who have experienced physical violence were 5.8 times more likely than men who did not
experience physical violence to suffer from serious mental illness. Some results from the
qualitative analysis may explain this latter finding. Although this view was not nationally
208
representative, an opinion leader in the Volta region mentioned that some men took to
drinking because of violence at home, which affected their mental capacity to work on the
farm or at any workplace.
4.6.3. Effects on work and daily activities
In line with the discussion in Section 2, the survey included a number of questions about
how each act of physical or sexual violence experienced in the 12 months prior to the
survey affected the daily lives of those who had experienced violence, in terms of their
absence from or ability to work, go to school or do domestic work, their ability to
concentrate on daily activities, their levels of confidence and on reporting feelings of living
in fear.
Thirty-two men and 78 women across the whole GFLHS 2015 reported consequences of
sexual violence. The majority of respondents affected by domestic sexual violence (90.6 per
cent of men and 87.2 per cent of women) reported no impact of sexual violence on their
daily activities. Ten women and four men felt that domestic violence had adversely
affected their ability to concentrate on work. Others (less than three individuals) reported
that they could not focus at school, were unable to concentrate on their domestic chores,
lost confidence in their own abilities and lived in constant fear.
The GFLHS 2015 also asked about the effects of physical violence on the daily lives of
individuals who experienced it. Fifty men and 192 women across the whole survey
reported effects of physical violence. The most prominent consequence for men was
inability to concentrate at work (four men). Among women, 39 were unable to concentrate
while doing their domestic work, 37 were unable to concentrate while at work, 30 lost
confidence in their own abilities, 22 lived in constant fear, and 21 missed work.
The qualitative data also showed that women became economically more vulnerable as a
result of domestic violence. This view was widely mentioned across all regions by
participants in both focus group discussions and in-depth interviews, who discussed the
extra burden placed by vengeful perpetrators as they refused to provide housekeeping or
chop money following other acts of violence. Reflecting this view, an opinion leader stated:
“The man will decide not to provide for the family. He will refuse to give chop money and
also stop paying the children’s school fees.” (IDI, opinion leader, Gbi Kpoeta, Hohoe
municipality, Volta region)
As a result, women often came to accept abusive relationships because of their economic
dependence on men. This economic motivation for the acceptance of violence was not
nationally representative but was more visible in highly patriarchal sites and areas with
greater levels of poverty in the northern regions, as illustrated by the following quote:
“The victims, especially the married women, get scared that when they report their husbands
to the police, they will be jailed and who takes care of them and their children.” (FGD,
Married women, Gbolo, Sagnerigu district, Northern region)
209
4.6.4. Consequences for children
Across all regions of Ghana, respondents in both focus group discussions and in-depth
interviews mentioned the wide-ranging consequences of domestic violence for children
and young people in the short and long term. A prominent effect discussed by respondents
was the psychological impact of domestic violence on children who are deprived of
parental care and socio-economic rights. This nationally representative view was well
illustrated by this quotation:
“Lack of peace in the family can cause the minds of children to retard. Their confidence level
is also reduced as they grow in domestic violence homes.” (IDI, opinion leader, Gbi
Kpoeta, Hohoe district, Volta region)
Respondent after respondent emphasised how care for, and attention to, children may be
compromised in homes that experience domestic violence. A respondent pointed to his
own experience, in a view shared by many in focus group discussions and in-depth
interviews across Ghana:
“It will affect the children emotionally when their daddy is groaning and their mum is
moody. I want to use my experience as an example. I had a little quarrel with my wife, the
child was crying, I was expecting my wife to go for the child and my wife was also expecting
me to go for the child, and because we were angry with each other, we didn’t know who
should go for the child. I just had to be a man and go for the child because she was not ready
to attend to the child.” (IDI, opinion leader, Garu, Garu Tempane district, Upper East
region)
Another way in which domestic violence may affect children was through the loss of their
mother. This effect was only mentioned in areas dominated by patrilineal culture, where
men retain possession of children upon separation, such as in the Upper East and Upper
West regions. This point was illustrated by an interview with an opinion leader:
“When it persists, the woman will seek separation or divorce and will become free from this
burden. When the woman leaves, the children will not have motherly care because in this
area, the children are considered to belong to the man.”
In areas of matrilineal inheritance in the Obuasi district of the Ashanti region, responses
typically pointed to women assuming responsibility for the care of children upon
separation, with children being deprived of paternal care and the economic support that
comes with it:
“Most often it is the children and the woman that suffer especially when they are left in the
care of an unemployed mother. Some men refuse to give house-keeping money for the use of
the mother and the children.” (IDI, opinion leader, Obuasi, Obuasi municipality,
Ashanti region)
210
Children growing up in abusive homes manifest their traumatic experience in school.
Many respondents from the field of education pointed out this link between abusive homes
and challenges in school:
“Yes. Even just this morning, which is a holiday, a woman drove her child to come to class.
Meanwhile the woman knows well that today is a holiday. To provide the child with just
GHc1 is a problem. So she must find ways and means to escape that burden of providing the
child with money, hence the decision to send her away from home. When the child comes to
school, as her friends are eating, she might be lucky to get something to eat. Even today, one
of my teachers gave a child GHc1 because the child was hungry. The moment she collected
the money, she set off for home. When I enquired where her mother was, she replied she was
at home. I enquired further and was told the parents are divorced and for that matter, when
she was coming to school, she wasn’t given any money. At times we are compelled to sacrifice
to feed some of these pupils, providing token money to these young children. Even we have
declared some of these children ‘HIPC’, because anything you ask, they won’t bring it. We
have about 8 pupils in this HIPC category in this school. At times too some of the teachers
buy dresses for these pupils. These are problems we go through. Thus some parents are not
treating their children well at all. This innocent child can be somebody in the near future.”
(IDI, opinion leader, Goaso, Ahafo-Ano district, Brong Ahafo region)
As a result of these experiences, children’s education is also affected, as they start:
“absenting themselves from school or neglect their studies altogether” (IDI, opinion leader,
Tomefa, Ga South district, Greater Accra region)
In addition, echoing a view mentioned across all ten regions:
“Some of the parents refuse to send their children to school. Some of the parents use the
children to do some kind of jobs, which children cannot do. […]Sometime too, threaten them
with dangerous weapons.” (IDI, opinion leader, Boakyekrom, Ahafo North district,
Brong Ahafo region)
These forms of neglect were discussed in particularly strong terms among focus group
discussions and in-depth interviews in areas with high levels of poverty, as this quote
illustrates:
“Some parents abandon their children to the point where some children go to school on empty
stomachs. Parents neglecting their children are a paramount situation here.” (IDI, opinion
leader, Jirapa, Jirapa district, Upper West region)
Children in turn may seek comfort in friends or resort to risky behaviours. The outcome in
the long term may be waywardness and social deviance among the youth, as most opinion
leaders involved in child protection services mentioned:
“The children are not spared under these conditions. They are also negatively affected. They
tend to look for happiness out of home due to the fighting and sometimes fall into bad
company. Some begin to smoke wee, drink and even steal. The girls also look for boyfriends
to support them because their fathers are not caring for them well. This often results in
211
pregnancies which have bad repercussions on their future or career.“ (IDI, opinion leader,
Gbi Kpoeta, Hohoe municipality, Volta region)
In essence, the results from the qualitative data highlighted how domestic violence
experienced in the home, and the resulting divorce or separation, had both short- and long-
term impacts on children’s educational, health, economic and social well-being. This is an
important finding because, as discussed in subsection 4.3.1, experiences of domestic
violence during childhood may themselves be reflect in vicious cycles of victimisation or
perpetration of violence in adulthood.
4.7. Use of and access to victims’ support services and institutions
This subsection addresses the final research question: Which institutional structures
support victims of domestic violence, and what proportion of victims are able to effectively
access available services? The analysis starts by examining the factors that may explain
why women and men who have experienced domestic violence seek (or not) external help
(subsection 4.7.1). It then looks at the public services and institutions available in Ghana to
support those affected by domestic violence (subsection 4.7.2), how many women and men
who participated in the study were aware of these services and institutions (subsection
4.7.3), and the levels of satisfaction with services among those who have used them
(subsection 4.7.4). The main aim of this analysis is to uncover information about the current
use and knowledge of and access to public services among individuals and communities,
which will be useful for future interventions aimed at preventing and mitigating domestic
violence in Ghana.
4.7.1. Seeking external help
A large international study found that most victims of domestic violence did not report
their abuse because domestic violence was often considered ‘normal’, or for fear of further
violence, losing their children or causing shame to the family (Garcia-Moreno et al., 2005).
Previous studies in Ghana had similar findings (Amoakohene, 2004). According to the
GFLHS 2015, 853 women and 419 men across the whole sample interviewed in the survey
reported having experienced at least one type of domestic violence in the 12 months
preceding the survey. Around 35 per cent of these women (297) and men (146) tried to
approach a person or organisation for help or support after experiencing violence. These
numbers were very similar to the DHS 2008 survey, which showed that 37 and 34 per cent
of women and men, respectively, who experienced violence had sought help.
Table 65 shows the share of women and men who sought external help after experiencing
any form of domestic violence. It reveals a wide variety of patterns among women and
men’s help-seeking behaviour:
Women were more likely to have sought help if they experienced physical violence
(52.8 per cent) or psychological violence (44.5 per cent) than for other forms of
violence.
212
Men were more likely to have sought help if they experienced psychological
violence (40.1 per cent) or economic violence (39.6 per cent) than for other forms of
violence.
Women were more likely than men to have sought help if they experienced sexual,
physical or psychological violence.
Men were more likely than women to have sought help if they had experienced
social or economic violence.
Table 65: Percentage of respondents who experienced domestic violence who reported
seeking external help (by sex)
Social
violence
Sexual
violence
Physical
violence
Psychological
violence
Economic
violence
Women 33.9 39.5 52.8 44.5 34.3
Men 35.6 35.1 34.3 40.1 39.6
N 531 110 242 487 508
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
When asked what made them seek help, the majority of women (58.5 per cent) and men
(53.3 per cent) answered that they “could not endure more”. The second most common
reason was the encouragement of friends and family (24.5 per cent for women and 28.5 per
cent for men), followed by the extent of injuries sustained (7.5 per cent for women and 8.0
per cent for men) (Table 66). These findings suggest that domestic violence is still largely
considered a private family matter: the GFLHS 2015 reported that 81.1 per cent of
respondents believed that “family problems should only be discussed with people in the
family”. Women and men who experienced domestic violence seemed to only resort to
external help in more extreme circumstances.
Table 66: Reasons for approaching external help after exposure to any form of
domestic violence by sex (in percentages)
Men Women
Encouraged by friends/family 28.5 24.5
Could not endure more 53.3 58.5
Badly injured 8.0 7.5
Afraid person/persons would kill me 0.6 1.7
The person/persons who did this to me
threatened or tried to kill me 0.9 0.5
Saw that children are suffering 0.0 0.7
Thrown out of the home 0.0 0.2
Encouraged by organisation 0.3 0.2
Other 7.4 6.3
N 146 297
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
213
Table 67 shows the reasons for not seeking help. The most common reason given was that
they did not think they needed help (44.1 per cent of men and 32.5 per cent of women). In
line with Garcia-Moreno et al. (2005) and the findings discussed above, 22.7 per cent of men
and 17 per cent of women did not seek help because they considered violence experienced
as ‘normal’ or ‘not serious’. In addition, 10.7 per cent of men and 16.4 per cent of women
did not seek help because they did not know who to ask – an issue that will be explored
further in the next subsections.
Table 67: Reasons for not seeking external help after exposure to any form of domestic
violence by sex (in percentages)
Men Women
Did not know who to ask 10.7 16.4
Person/organisation too far 0.2 0.1
Nobody can help me 6.9 7.6
I don’t need help 44.1 32.5
If my partner finds out it will cause trouble 0.1 1.7
If other family member finds out it will cause trouble 0.7 2.0
I don’t trust anyone 2.4 5.0
Violence normal/not serious 22.7 17.0
Embarrassed/ashamed 2.5 4.5
Afraid would not be believed 0.5 0.5
Afraid would be blamed 0.8 1.2
Afraid would end relationship 0.0 1.4
Afraid would lose children 0.0 0.3
Bring bad name to family 3.0 3.0
Other 4.6 5.7
Don't know or refused to answer 0.8 1.1
N 268 550
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
The qualitative data allowed the research team to explore the quantitative results further.
Data collected through focus group discussions and in-depth interviews in all ten regions
in Ghana revealed that distrust of, and poor experiences with, state authorities contributed
to individuals not seeking help. A large number of the respondents mentioned three key
challenges: the costs of the services, lengthy or ineffective delays in the handling of cases,
and corruption. Corruption, in particular, was said to affect access to justice when survivors
were given bribes and when perpetrators paid bribes or used their networks of influence to
avoid prosecution. This view was well illustrated by the following statement:
“There are times the assemblymen go to the police station to ask for the discharge of a culprit.
Normally when they have some small money on them they give it to the police in charge and
the matter will be dropped.” (FGD, married/living together men 18–50, Gbawe, Ga
South district, Greater Accra region)
214
There was a widespread view that particularly the police and the courts collected money
from those who approached them, but did not deliver appropriate care in return. This view
is exemplified by this respondent:
“For the police, they will collect money from you and would not help you solve the
problem.”(FGD, unmarried youth, Sagnarigu district, Northern region)
An opinion leader interviewed in another region added:
“People only take their case to the chief because they cannot afford the monies the police will
demand from them.” (IDI, opinion leader, Wuling, Jirrapa district, Upper West region)
Respondents also mentioned that victims were often sent back by the police to their families
to find a solution, and prosecution was discouraged:
“If a police staff knows the victim, they will ask you to take the matter to the family level for
solution.” (FGD, married/living together men 18–50, Gbawe, Ga South district,
Greater Accra region)
Concurring with this widespread perception of domestic violence as a private matter, even
within public services set up to deal with domestic violence, the majority of respondents in
the GFLHS 2015 who sought help did so in the first instance by approaching family
members or a friend (66.2 per cent of men and 77.6 per cent of women). The second source
of help was the police (sought by 14.6 per cent of men and 9.0 per cent of women as a first
point of contact) (Table 68).
Table 68: First choice of help (in percentages)
Men Women
Family member or friend 66.2 77.6
Health centre, hospital or other health
institution 1.4 1.2
Police 14.6 9.0
Queen mother 0.9 0.7
Domestic Violence and Victim Support Unit 0.6 0.7
NGO, CSO or social worker 0.6 0.2
Lawyer or court 0.6 0.7
Traditional leaders 4.9 3.0
Religious leaders 2.3 3.5
Community group 2.6 1.2
Other 4.3 2.3
Don't know or did not answer 0.9 0.0
N 145 297
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
The qualitative data revealed, however, that the preference of study participants for
traditional or formal legal ways of addressing domestic violence was often confounded by
215
underlying deficiencies in the justice system that, combined with structural inequalities,
may have prevented victims of domestic violence from accessing it (see also Brickell, 2015a,
2015b). As stated earlier, the 2007 Domestic Violence Act encourages mediation of all types,
but its reconciliatory approach brings some challenges. In addition to the challenges
discussed above, poorer victims and those worried about personal social stigmatisation
and reputational loss for the whole family may have preferred not to access formal justice
processes. Also, the study found that respondents were more likely to withdraw a case
from prosecution and seek customary mediation when the family or the community
determined that the victim was:
”to blame … for being the cause of the problem, when the perpetrator shows remorse [and/or]
when the family thinks about stigmatisation.” (IDI, opinion leader, Boakyekrom, Ahafo
Ano district, Brong Ahafo region)
Furthermore, respondents in focus group discussions said:
“Family and community may discourage prosecution if it is minor case like verbal assault”
or when prosecution could “lead to divorce” (FGD, married/living together women,
Garu, Garu Tempane district, Upper East region)
And, in many cases:
“…the family would convince the victim not to report the case or follow it up for …
prosecution of the perpetrator because the offender may be an opinion leader in the
community or the breadwinner of the family. Also if they like the offender, then they would
discourage his prosecution” and since “many cases of domestic violence that get prosecuted
end up in divorce, … if the couple love each other [they] would prefer to work out their
differences, the family would intervene to make sure that the union is not dissolved and so
prevent divorce and family feuds; the family and some members of the community always
try to discourage prosecution of such cases.” (IDI, DOVVSU, Gbawe, Ga South district,
Greater Accra region)
As these responses indicate, the rights of the victims are not always placed at the forefront
of customary mediation processes. Customary mediation may be strongly desired and
provide an important reconciliatory cost/time-effective approach in the Ghanaian social
context. Yet, the rights of the victim may often become submerged by the social and
economic needs of the family or community, and by prevalent gender norms. This situation
may weaken victims’ rights, while encouraging perpetrators to act with impunity.
4.7.2. Public services and institutions
The Government of Ghana has adopted a multi-institutional approach to support victims
of domestic violence after the enactment of the 2007 Domestic Violence Act. The primary
institutional response is the Domestic Violence Victim Support Unit (DOVVSU) of the
Ghana Police Service. DOVVSU acts in partnership with the Department of Social
Development, the International Federation of Women Lawyers (FIDA) and the Legal Aid
Scheme, and collaborates with the health system to provide comprehensive support to
216
victims of domestic violence. To better understand how these services currently function,
the research team conducted several observation exercises and interviews within DOVVSU
units (as detailed in Appendix C), which gathered detailed information on: (i) the facilities
available to and protocols followed by DOVVSU personnel; (ii) levels of training of
DOVVSU personnel; and (iii) the extent of the collaboration between DOVVSU and public
health services.
Facilities and protocols. The 2007 Domestic Violence Act makes provisions for police
assistance to victims and perpetrators of domestic violence. These include receiving
complaints, offering advice about victims’ rights and about services available, helping
victims to find a safe refuge when required, and providing assistance with free medical
treatment and medical evidence for possible prosecution. The interviews and observations
conducted during the study revealed that none of the DOVVSU units visited had all the
necessary facilities to offer these support services effectively. In particular, they lacked
private rooms, shelter facilities, counselling services, protection services or prosecution
advice. DOVVSU personnel interviewed in the Upper West region explained that this
situation is common across all ten regions:
“We are handicapped so if you people could help us get fresh accommodation, it will help us
very well in exercising our duties. We also need vehicles and other logistics to help us in our
line of duty.”
The research team’s visits to DOVVSU offices also revealed that, to date, the Ghana Police
Service does not provide victims’ shelters at the national or district level. As mentioned in
one of the interviews conducted with DOVVSU personnel in the Western region:
“DOVVSU have a lot of challenges […]We have issues with accommodation and cannot
house victims when the need arises. If DOVVSU had enough office space, it would have been
convenient for some victims to be brought here.”
The only CSO currently seeking to provide such a service, the Ark Foundation, receives no
support from the government.32 Most DOVVSU units consisted mainly of personnel with
limited office space. The few which had their own offices did not always have computers
to keep their records.
DOVVSU officials in all offices visited were found to be working under difficult conditions
and, sometimes, paying for the medical or transportation bills of victims themselves. The
needs of DOVVSU emerge in this poignant observation by the team at one of the DOVVSU
units included in this study:
“I saw that the police worked under difficult conditions ... they have a cage-like place near
complaints which had two compartments, men who were detained were placed in the inner
cell while women were placed in the outer one. […] I saw that DOVVSU needed a lot of
support […] as the interviewed officer told me, […] DOVVSU did not even have a car to
effect arrests. They had often had to use their little salary to foot transportation without
32 Information obtained in interviews with Mr Sandogo, Police Social Services, Director of the Ark
Foundation and National Coordinator of the Domestic Violence Coalition in Ghana.
217
getting reimbursed. The winds often blew water into their offices as it rained because the
louvre blades in the windows were either broken or not functioning anymore. This affected
the work of the officials, as they sometimes have to move from their offices taking files along
with them. Lastly, I must say this Officer was very helpful. […] The interviewed Officer was
very friendly in nature and as I was interviewing him, we were constantly interrupted by
people coming to call him out to assist them with their case at the unit.”
Other detailed observation exercises revealed that, despite being overwhelmed with their
workload and often lacking adequate facilities, the conduct of the police while dealing with
cases of domestic violence was very professional, with care being taken to file documents
appropriately (within infrastructure constraints) and to provide as much privacy and
support as possible to those seeking their help. However, the facilities available varied
considerably across regions, as follows:
Greater Accra region, Gbawe, Ga South district: There was no private room for
victims to report their cases comfortably. All the offices visited, except the
Commander’s office, looked cramped, with very little room for people to move. At
the DOVVSU desk, there were over five personnel in a very small room. There was
no fan, so the room was very hot. The room looked like an extension to the main
building; from the outside, it looked like a cage. The seats in the offices were worn
out, and many of them were old benches. Files were well stored away from the
public.
Eastern region, Kukurantumi-Koforidua: DOVVSU officials were using four
different offices. Another office, called the secretariat, was used to store all files
related to domestic violence. There was a private room where victims could sit to
file complaints.
Volta region, Gbi Kpoeta, Hohoe municipality: The police station had DOVVSU
personnel on site, but there was no separate structure for the DOVVSU unit to use.
There was only a small office that was used by the officer-in-charge. The rest of the
DOVVSU officials shared the same office with the other police officers. They
provided counselling in the same office as other police officers who could listen to
the victims’ stories. There was very limited privacy.
Northern region, Kalpohin, Sagnerigu district: The office for the entire unit
consisted of two small rooms, one of which was used by the head of the unit. The
other office was shared by about ten officers. There was limited space for people to
move about. There was no private room for victims, and reports had to be made in
front of others. However, files were kept safely out of sight.
Upper West region, Wa municipality: Office space was very small, with files stored
in a locker which looked quite old. There was no private place where victims could
report their cases. Victims had to sit in an open place where anybody could hear
them. The office did not have enough chairs and desks for the police personnel and
victims. There were no vehicles for police personnel to use in effecting arrests or to
follow up cases under investigation. The research team observed that DOVVSU
officers were not in uniform, which made them more approachable to community
members. They were very friendly with victims who approached their offices.
218
Upper East region, Garu Tempane district: The district did not have a DOVVSU unit
or staff. Interviews were conducted with district police officers in their normal line
of duty. Files were stored in a separate out-of-sight room. There was no privacy,
with people being first directed to the charge office, where an officer registered their
complaint. After listening, victims were referred to an investigator, who took their
statement. Victims were offered a seat in front of the investigator in a private room,
where he would write a detailed statement. A medical form was issued afterwards
if necessary. The victim was invited to lead the police in the arrest of the perpetrator.
If witnesses were available, victims were asked to provide them and lead the
investigator to the scene of the crime. Victims were expected to avail themselves in
court to testify.
Brong Ahafo region, Goaso, Ahafo North district: There was no building or office
which catered solely for domestic violence victims. All reports were made at the
charge office.
Central region, Cape Coast district: DOVVSU had its own separate office space, but
it was in a poor state of repair. No private room was observed.
Ashanti region, Obuasi municipality: Files were stored in a cupboard in a restricted
room at the police station. Privacy was ensured, and the procedure for filing a
complaint was the same as described above for the Upper East region.
Western region, Tarkwa district: The DOVVSU unit had two rooms. The DOVVSU
person was also the head of the police division and had to balance different duties.
The two rooms afforded limited privacy: the first room functioned also as a waiting
room and a reception area, whereas the second room was the head’s office. It was
also where files were kept.
Training of DOVVSU personnel. The establishment of DOVVSU units was intended to be
accompanied by the training of personnel to respond to domestic violence cases. The large
majority of the police personnel interviewed mentioned that they had not received
specialised training to provide help to victims of domestic violence. Neither had they
received training on gender-based violence or domestic violence, or specialised training in
listening skills to better elicit information from those seeking help. The major formal
training by the Ghana Police Service on gender-based violence took place in 2004. With the
help of UNICEF, a training manual was recently prepared to make training in gender-
based violence a basic part of the curriculum of the Police Training School; it should soon
be more widely used. DOVVSU officers were also observed to have received limited
training in information technology, and most district offices had no digital means or
training to record or access local records.
Collaboration between DOVVSU and public health services. There is close collaboration
between DOVVSU and the district hospitals which provide treatment to victims of
violence. This collaboration does not, however, include support services to victims of
violence. This is because, to enable prosecution of assault cases, medical reports have to be
completed by the medical doctor who treated the victim of domestic violence. But these
medical services are not free of charge. Many doctors request ‘unofficial’ fees from victims
219
to cover the cost of writing the medical report and the possibility of having to go to court
to explain it. This makes prosecution inaccessible to many victims, especially low-income
women. One opinion leader interviewed in the Ashanti region expressed this view, which
was common across all regions in Ghana:
“DOVVSU is really responsible for addressing domestic violence issues, but one thing
missing in their work is a centre for protection, where they will keep the victim before they
carry on with investigations. This is an issue. Secondly, when it comes to issuing a medical
report and they send it to the hospital, the doctors ask for a fee before they sign the report.
Most of these victims are vulnerable people, and they are not able to pay such fees and this
is a worry to us.”
Another interview with a medical doctor who is a domestic violence activist revealed that
many medical doctors were not trained on how to write medical reports for prosecution,
which may sometimes give lawyers of perpetrators the upper hand in reinterpreting the
reports to the benefit of their clients. Also, district hospitals do not have specialised
services, such as emergency responses to victims of violence, or priority systems for victims
of abuse unless they were brought into the hospital as normal emergency cases.
In summary, while DOVVSU is lodged within an ideal multi-institutional framework
within which to ensure effective support to victims of domestic violence, it is currently
inadequately equipped to make operational the 2007 Domestic Violence Act. At the time
of this study, DOVVSU faced the same institutional constraints that existed in the police
service prior to legal reform (see Cusack and Manuh, 2009; Ellsberg and Heise, 2005).
4.7.3. Knowledge about support services and institutions
Study respondents across all regions of Ghana showed awareness about the public services
and institutions that deal with domestic violence (Table 69). DOVVSU, however, was still
largely known by its previous name – the Women and Juvenile Unit (WAJU). The few
districts where knowledge about the public support system was more limited (in the
northern regions) were also districts where DOVVSU services were not available. In those
districts, respondents typically stated they had neither heard, seen nor used any such
public facilities to resolve domestic conflicts. But in districts where DOVVSU was
established, respondents generally included DOVVSU in their list of known public
institutions.
Table 69 provides an overview of the knowledge exhibited by the GFLHS 2015 respondents
about a range of available public services. The level of knowledge about public services
and organisations was quite high among all respondents and across all different socio-
economic groups and regions. With respect to DOVVSU, 84.9 per cent of all respondents
knew about their local units. This level of knowledge was lower among women (83.7 per
cent) than men (86.6 per cent), those living in rural areas (80.3 per cent) than those in urban
areas (89.3 per cent), those living in the Central, Volta and Northern regions (respectively,
74.2 per cent, 53.6 per cent and 67.4 per cent, compared, for instance, to 99.0 per cent in the
220
Western region) and those who had no education (71.9 per cent, compared to 98.0 per cent
for those with higher education) and were in the lowest asset quintile (70.1 per cent,
compared to 95.1 per cent for those in the top asset quintile).
221
Table 69: Respondents’ knowledge about public services (in percentages)
1 2 3 4 5 6 7 8 9 10 11 12
Sex
Male 99.9 98.9 86.6 53.8 78.9 94.9 82.6 94.7 98.4 99.6 98.7 97.9 Female 99.6 99.2 83.7 53.8 74.3 92.8 81.9 94.8 97.6 99.4 98.3 96.9
Age group (years)
15–19 99.2 99.5 80.5 52.7 71.9 92.7 78.2 95.2 99.0 99.7 98.2 97.5 20–24 99.1 98.7 85.8 56.5 78.1 94.5 82.0 94.6 96.8 98.8 98.0 97.1 25–29 99.9 98.8 87.7 53.9 78.0 93.5 84.3 94.8 97.5 99.2 98.2 97.1 30–39 99.6 99.1 86.5 55.6 76.7 93.3 83.0 94.4 97.9 99.6 98.4 96.7 40–49 100.0 99.2 81.9 50.6 73.4 94.0 81.0 94.5 97.9 99.8 98.8 98.0 50–60 100.0 99.2 83.6 52.4 76.4 93.7 82.0 95.7 98.9 99.7 99.1 98.2
Employment
Self-employed 99.7 99.0 82.3 51.9 73.9 92.5 79.2 95.0 98.6 99.5 98.2 97.2 Employed 99.8 98.8 93.0 59.6 82.6 96.1 90.3 93.6 95.6 99.1 98.7 97.2 Not working 99.6 99.4 85.7 54.6 77.1 94.7 84.0 95.1 97.9 99.8 98.9 97.6
Marital status
Never married 99.6 99.6 82.2 54.9 75.3 93.9 81.4 94.3 98.6 99.6 98.0 97.4 Married or living together 99.7 98.9 84.8 53.9 76.4 93.2 81.5 94.6 97.7 99.3 98.4 97.3 Divorced/separated/widowed 99.8 99.2 86.7 52.8 75.6 94.9 85.0 95.6 98.2 99.9 99.0 97.5
Residence
Urban 99.8 99.1 89.3 56.8 79.6 95.8 87.8 94.6 96.6 99.6 98.9 97.2 Rural 99.6 99.0 80.3 50.7 72.5 91.4 76.4 95.0 99.3 99.4 97.9 97.5
222
1 2 3 4 5 6 7 8 9 10 11 12
Region
Western 100.0 100.0 99.0 54.3 80.8 99.0 98.0 98.4 99.0 99.6 99.4 99.2 Central 99.5 98.4 74.2 25.8 54.0 96.5 84.7 96.5 98.8 99.5 99.5 97.7 Greater Accra 100.0 98.9 93.9 55.7 79.8 95.1 92.8 91.1 93.7 99.8 99.5 98.1 Volta 99.3 98.6 53.6 35.6 58.2 84.1 59.1 94.7 98.4 99.3 99.1 98.4 Eastern 100.0 100.0 96.0 72.4 92.4 99.0 98.9 100.0 100.0 100.0 100.0 99.8 Ashanti 99.9 99.5 92.6 70.1 83.3 97.1 87.9 99.1 99.5 99.8 99.5 98.6 Brong Ahafo 99.8 97.9 85.7 66.6 74.9 85.5 75.3 86.4 95.3 98.3 90.9 85.3 Northern 98.6 99.4 67.4 34.8 69.1 92.2 62.3 91.8 99.2 99.2 97.5 97.9 Upper East 100.0 97.7 84.1 21.5 72.9 85.0 51.9 91.1 99.1 99.1 99.1 99.1 Upper West 99.2 98.5 80.0 80.8 86.9 89.2 69.2 95.4 99.2 99.2 99.2 100.0
Education level
None 99.3 99.1 71.9 41.6 68.9 88.4 69.6 93.0 99.0 99.4 98.5 97.9 Primary 100.0 98.9 81.7 49.3 68.5 92.9 78.4 95.3 98.5 99.3 98.3 96.8 Middle/JSS/JHS 99.6 99.3 87.2 56.0 75.5 94.6 85.3 95.8 98.3 99.7 98.4 97.2 Secondary 100.0 99.3 90.0 61.1 84.8 96.2 86.0 94.3 97.4 99.3 98.7 97.3 Technical 100.0 97.9 91.0 56.9 77.8 95.1 89.6 95.8 96.5 99.3 99.3 97.2 Higher 99.8 98.2 98.0 65.0 91.1 97.4 94.1 94.1 94.7 99.4 98.2 97.4
Asset quintile
Lowest 99.3 98.7 70.1 43.2 65.9 88.2 67.6 92.9 99.0 99.5 98.0 97.7 Second 99.7 99.3 83.6 51.4 72.6 93.6 80.8 95.0 99.1 99.5 98.5 97.7 Middle 99.8 99.1 88.4 54.6 76.5 94.3 84.3 95.3 97.8 99.5 98.4 96.9 Fourth 99.8 98.8 90.4 59.1 82.1 95.6 89.8 96.4 96.6 99.4 98.8 97.6 Highest 100.0 99.3 95.1 63.4 86.7 97.4 91.1 94.3 96.6 99.5 98.4 96.3
Total 99.7 99.1 84.9 53.8 76.1 93.6 82.2 94.8 97.9 99.5 98.5 97.3
Source: Ghana Family Life and Health Survey (GFLHS) 2015. Notes: Service providers are 1 = Health providers, 2 = Police, 3 = DOVVSU, 4 = Shelter, 5 = NGO, CSO or social worker, 6 = Lawyer or member of Court,
7 = Legal aid group, 8 = Queen mother, 9 = Traditional leaders, 10 = Religious leaders, 11 = Religious group, 12 = Community group
223
The quantitative survey showed that despite good knowledge about public services in
general, and DOVVSU in particular, there was some room for improvement in terms of
awareness about the existence of a law against domestic violence in Ghana. According to
the GFLHS 2015, 75 per cent of respondents in urban areas and 61 per cent in rural areas
knew there was a law against domestic violence. There was also a large variation across
regions in terms of awareness about the Domestic Violence Act. For instance, only between
10.7 and 12.1 per cent of those in the Upper East, Upper West and Northern regions knew
about it.
4.7.4. Levels of satisfaction about support services and institutions
The information collected in the study has revealed that satisfaction with the public
primary response institution against domestic violence, DOVVSU, was low in comparison
with other means of support. However, very few respondents answered this question, as
not many individuals used these services. Table 70 provides an overview of satisfaction
levels with services, such as health providers, police and DOVVSU units, as well as with
traditional avenues for conflict resolution, such as traditional and religious leaders and
community groups.
Table 70: Satisfaction levels with public services (number of respondents)
Completely
satisfied
Somewhat
satisfied
Somewhat
dissatisfied
Completely
dissatisfied N
Health centre, hospital,
other health provider 10 1 0 1 12
Police 60 17 11 17 105
DOVVSU 2 1 1 2 6
Traditional leaders 19 10 2 4 35
Religious leaders 20 4 3 2 29
Community group 9 2 3 2 16
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
The qualitative data revealed that the reasons for the lower levels of satisfaction with
DOVVSU were mostly related to the direct and indirect costs of services, including legal
services. What bothered people was not just the financial cost but also the cost of secondary
victimisation due to lengthy procedures and additional payments:
“At the police facility, they waste a lot of time and always make demand for money so I am
really not satisfied with their service. I am however satisfied with help at health centres
because they are efficient and sometimes even treat victims of domestic violence without
referring it to the police to avoid prosecution” (IDI, opinion leader, Sagnerigu district,
Northern region)
224
Families do not always have the funds to pay for the legal costs of prosecution, which puts
them in a difficult situation:
“The main challenge has to do with the finance. Some of the clients, especially in cases of
domestic violence cannot afford the fees for their legal representation and often, we have to
take up these cases pro bono.” (IDI, opinion leader, Gbolo, Sagnerigu district, Northern
region)
GFLHS 2015 respondents were asked whether they had to pay for services when
approaching any relevant institution. The results showed that about one quarter of service
users had to pay for them. Twenty per cent of respondents who used police services paid
a fee, and about 6 per cent of respondents paid a bribe. This situation has prevented many
from accessing the services they need, and has created considerable constraints to those
who try to help and support women and men affected by domestic violence in Ghana.
225
5. Summary and discussion
This study has provided an in-depth analysis of the attitudes to and incidence,
determinants and consequences of domestic violence in Ghana, through the collection of a
rich set of quantitative and qualitative data. This section summarises the main findings of
the study in relation to each research question.
5.1. Incidence of domestic violence in Ghana
Over a quarter (27.7 per cent) of women and 20 per cent of men in Ghana
experienced at least one type of domestic violence in the 12 months prior to the
survey.
The most common form of domestic violence reported by women was economic
violence (12.8 per cent), followed by social violence (11.6 per cent), psychological
violence (9.3 per cent), physical violence (6.0 per cent) and sexual violence (2.5 per
cent).
The incidence of domestic violence was generally lower among men than among
women. The most common form of domestic violence experienced by men was
psychological violence (7.9 per cent), followed by social violence (7.7 per cent),
economic violence (7.3 per cent), physical violence (2.1 per cent) and sexual violence
(1.4 per cent).
The analysis of the qualitative and quantitative data suggested that sexual violence
was largely considered a private matter. The results above may, therefore,
underestimate the true extent of this form of domestic violence.
The GFLHS 2015 is not comparable to prior surveys of domestic violence in Ghana
because it collected data on broader and more in-depth definitions of domestic
violence than other surveys. It is, however, possible to compare the incidence of
physical domestic violence experienced by women and men in the 2008 DHS and in
the 2015 GFLHS. The results show, when using comparable categories of domestic
physical violence,33 that domestic physical violence among women decreased from
17.2 per cent in 2008 to 10.3 per cent in 2015. Domestic physical violence among men
decreased from 12.7 per cent in 2008 to 11.2 per cent in 2015.
There was a dominant view that domestic violence was largely perpetrated by men
against women. The idea of physical violence committed by women against men
was found to be ‘funny’ and sometimes ‘unimaginable’. The qualitative and
33 Please note that the estimates below for the incidence of domestic physical violence in 2015 do not match
the estimates provided in Table 26. This is because the DHS 2008 considers domestic violence to include some
categories of perpetrators that are not part of the definition of domestic relation used in this study (or in the
Domestic Violence Act 732).
226
quantitative data found, however, several instances of domestic violence committed
by women against men. For instance, 12 per cent of physical domestic violence and
37 per cent of psychological domestic violence against male household heads was
perpetrated by their spouse or (former) partner. Among the participants in the
GFLHS 2015, 789 women and 563 men reported having committed at least one form
of domestic violence, but these results must be interpreted with care due to potential
under-reporting and reporting biases.
5.2. Attitudes towards domestic violence in Ghana
Respondents overall opposed all forms of domestic violence. However, there were
important caveats as to what was perceived to constitute domestic violence.
Respondents considered that some of the definitions of domestic violence used in
the study (and in Act 732) were not violence but were, rather, accepted social norms,
defence mechanisms or ways of resolving disputes. Only physical violence between
partners was consistently perceived as violence. This is an important finding
because it suggests that violence is experienced, conceived and articulated in
different ways along a spectrum that locates some forms of violence (such as
physical violence) as the most severe, and other forms of violence (for instance,
controlling behaviour by women) as a way of managing unequal power in domestic
relationships.
Psychological violence was generally perceived as a means of defence and a coping
strategy – for women in particular – to ward off physical violence or to claim some
autonomy or agency in contexts where they may not have the ability to exert much
power beyond the use of words (insults), their body (withholding sex) or their
housework (not cooking).
Physical violence was largely considered to be unacceptable across Ghana, but some
acts of domestic physical violence were sometimes deemed acceptable, depending
on the context. There was, in particular, acceptance of wife-beating as a punishment
against disobedience and neglect of children: 11.2 per cent of men and 17.2 per cent
of women mentioned that wife-beating was acceptable if the wife had disobeyed the
husband, and 10.2 per cent of men and 16.5 per cent of women agreed that it was
acceptable if the wife neglected the children.
Attitudes of support for wife-beating have, however, become less pervasive in
Ghana since 2008: the proportion of women justifying wife-beating decreased from
36.6 per cent in 2008 to 23.1 per cent in 2015, while the proportion of men condoning
wife-beating decreased from 21.8 per cent in 2008 to 13.8 per cent in 2015.
Physical violence against children was generally viewed as unacceptable, but was
described in some discussions, particularly in the Northern and Eastern regions, as
a form of parental education to ‘steer them in the right direction’. There has been,
227
however, a significant reduction in attitudes condoning physical violence against
children: in 1998, 8 per cent of respondents thought it was never acceptable to beat
a child (Coker-Appiah and Cusack, 1999), whereas in 2015, this number rose to 25.3
per cent among men and 18.8 per cent among women.
Women were on average more likely than men to find wife-beating acceptable: 23.1
per cent of women and 13.8 per cent of men condoned wife-beating. Furthermore,
65.3 per cent of women and 56.2 per cent of men fully agreed that women were to
blame for rape if they wore revealing clothes. These findings are in line with
previous literature and reflect three facts. First, that the above forms of domestic
violence are still considered acceptable. Second, victims tend to internalise social
norms once they are in abusive relationships. Third, even when not victims, women
may internalise gender norms that hold them responsible for their experiences of
gender-based violence, rather than placing that responsibility on men or on the
broader system that reinforces gender inequalities.
5.3. Determinants of domestic violence in Ghana
The study revealed that age and intergenerational factors were strong determinants of
domestic violence in Ghana:
Young women and men (15–19 years old) were substantially more likely to have
experienced domestic violence than other age groups. Women aged 15–19 years
were between 1.3 and 4 times more likely to experience any form of domestic
violence than, for instance, women aged 30–39 years. Differences in the incidence of
domestic violence across age groups were similar for men, with the exception of
sexual violence, which was more prevalent among men aged 20–24 years. This is in
contrast with the 2008 DHS, which reported very similar incidence across age
groups, but in line with the 1998 study by Coker-Appiah and Cusack (1999).
Exposure to violence in childhood was found to be strongly related to the likelihood
of an individual being a victim of violence in adulthood. Women who witnessed
any form of domestic violence as children were between 1.3 and 3 times more likely
to report domestic violence than women who did not witness violence during their
childhood. Men who witnessed social, psychological or economic violence in their
homes as children were between 1.2 and 2.2 times more likely to have experienced
these forms of violence in adulthood than other men. This is a common finding in
domestic violence research, including the study by Tenkorang et al. (2013) on
domestic violence in Ghana.
Exposure to violence in childhood was also found to be related to the likelihood of
an individual being a perpetrator of violence in adulthood: about 28 per cent of
women and men who admitted committing social violence witnessed social
violence as children, compared to 14–15 per cent of those women and men who did
not experience violence as children. Among men who witnessed sexual violence as
children, almost 30 per cent committed sexual violence over the 12 months prior to
the survey, compared to 2.6 per cent of men who did not experience this form of
228
domestic violence during their childhood. Across psychological, physical and
economic violence, rates of perpetration are between 2 and 3 times higher for
women and men who witnessed domestic violence as children than among
individuals who did not experience domestic violence during their childhood.
Taken together, these findings suggest that Ghana may experience vicious cycles of
domestic violence reinforced by the fact that: (i) young people are more at risk of
domestic violence; (ii) experiencing domestic violence at a young age will
considerably increase the odds of experiencing domestic violence later in life; and
(iii) experiencing domestic violence at a young age will considerably increase the
odds of the individual perpetrating violence later.
Socio-economic factors were reported to be among the most important determinants of
domestic violence in Ghana, albeit in complex ways, as follows:
Low asset levels were found to be strong determinants of psychological and
economic violence among women, but high asset levels were associated with higher
levels of social and sexual violence among men.
Tensions and fights over money or property were found to be leading determinants
of all forms of domestic violence across all regions.
Women with no education were between 0.5 and 2.5 times less likely to experience
domestic violence than women with some level of education. More educated men
were around 3 times more likely to experience economic and sexual violence than
less educated men. This result may be due to the fact that women and men with
more education may also be more likely to report domestic violence, as emphasised
in previous studies conducted internationally and in Ghana.
Factors such as poverty, unemployment and economic shocks and associated
tensions were identified as important determinants of domestic violence. However,
employment status was not found to be a particularly significant determinant of
domestic violence, with some exceptions: women who were not working were 43
per cent more likely to have experienced economic violence than self-employed
women, while men who were employed or not working were, respectively, 2 times
and 2.5 times more likely to have experienced economic violence than self-
employed men.
Sexual violence was found to be more pronounced among individuals with higher
levels of education and socio-economic status, particularly men.
The study did not find a clear relationship between marital status and domestic violence.
However, the following findings are noteworthy:
Divorced, separated or widowed women were twice as likely to have experienced
physical violence than women who were never married.
Married women or divorced, separated or widowed women were, respectively, 1.8
times and 2 times more likely to have experienced economic violence than non-
married women.
Married or divorced, separated or widowed men were, respectively, 6 times and
almost 8 times more likely to have experienced sexual violence than non-married
229
men, and almost 3 times and 1.7 times more likely to have experienced social
violence than men who were never married.
Women in polygamous relationships were almost twice as likely to experience
psychological violence than women who were never married. These women were
also 3.5 times more likely to have experienced physical violence and 4.2 times more
likely to have experienced economic violence. Men in a polygamous relationship
were also more likely than men in monogamous relationships to experience sexual
or social violence.
Domestic violence sometimes followed women’s refusal to have sex with their
husbands, but this form of marital dispute was often reported to be used as a way
to retain some degree of power and autonomy in the face of other forms of violence,
especially economic violence.
There was no clear statistical relationship between geographical factors (urban/rural and
regional) and the incidence of domestic violence once the statistical analysis controlled for
other factors. This was probably because domestic violence was not necessarily related to
geographical characteristics per se but, rather, to the incidence of lower levels of
development in certain regions.
There was an association between the incidence of domestic violence and perceptions of
gender roles, patriarchal norms and masculinity. The results were particularly striking in
the qualitative data, which showed domestic violence to be closely determined by strong
gender norms that perpetuate gender inequalities. The survey also showed the following:
Women who exhibited higher levels of acceptance of wife-beating were 14 per cent
more likely to have experienced economic violence than other women. This is,
however, not a causal effect and may well suggest that victims of violence
internalise and accept gender norms that perpetuate or justify domestic violence.
Women who expressed a higher acceptance of women’s sexual autonomy were 16
per cent more likely to have experienced psychological or social violence, 17 per
cent more likely to have experienced physical violence, and 11 per cent more likely
to have experienced economic violence than women with a lower acceptance of
women’s sexual autonomy. This finding suggests that women who are more likely
to challenge accepted gender norms may be at risk of domestic violence because
they may be punished for not complying with established norms of behaviour.
Men who expressed stronger patriarchal gender norms were 26 per cent less likely
to report social violence than men who did not accept such norms.
Men who displayed higher levels of tolerance towards wife-beating were 21 per
cent, 47 per cent and 30 per cent more likely to have experienced psychological,
physical or economic violence, respectively, than men who reported lower levels of
tolerance towards wife-beating. This result may potentially reflect retaliation by
women who were beaten by their husbands.
Men who exercised more decision-making power within their households were 34
per cent less likely to experience social violence than men who exercised lower
levels of decision-making within the household.
230
The study found that exposure to criminal forms of violence in the community (theft or
robbery, vandalism, arson, kidnapping or abduction, extortion, bullying, political violence,
highway attacks, harassment, rape or defilement) was significantly associated with higher
incidence of domestic violence among men and women.
The odds of victimisation among women increased by 77 per cent for psychological
violence, 49 per cent for physical violence and 57 per cent for economic violence
when the respondent lived in a community that experienced at least one form of
violence. The effect was stronger for sexual violence and social violence, for which
the odds of having experienced this violence were 3.6 times and 1.9 times higher,
respectively, for women exposed to violence in their communities than for women
in communities that did not experience other forms of violence.
Men who reported some violence in their communities were between 2 and 3 times
more likely than men living in more peaceful communities to report having
experienced any form of domestic violence.
These findings emphasise the importance of taking into consideration factors
beyond the individual and the household when studying the determinants of
domestic violence.
Overall, the determinants of domestic violence varied considerably among types of
violence, emphasising the need for domestic violence legislation in Ghana to continue to
distinguish between different forms of violence (as in Act 732). Only young age, exposure
to domestic violence as a child and high levels of violence in the community were common
determinants of domestic violence across all types.
5.4. Consequences of domestic violence in Ghana
The GFLHS 2015 revealed adverse consequences of domestic violence on physical
health, including injuries and illness: 43.8 per cent of women who had been
physically assaulted declared having been ill in the 30 days prior to the survey – a
rate that was a third higher than among women who had not been physically
assaulted (31.2 per cent) – and 42.3 per cent of women who had experienced
psychological violence reported ill health, compared to 30.9 per cent of women who
had not experienced this form of domestic violence. The effects of domestic violence
on physical health were observed almost exclusively among women.
The results showed a very strong correlation between exposure to domestic violence
and mental health, particularly among women: women who reported serious
mental illness were 26.7, 24.2, 21.9, 19.3 and 26.3 per cent more likely to have
experienced, respectively, psychological, physical, sexual, social or economic
violence than women who did not report serious mental illness. This finding is in
line with previous studies in Ghana (Amoakohene, 2004; Adu-Gyamfi, 2014). Only
physical violence had an effect on men’s mental health status.
231
Domestic violence was found to have adverse qualitative consequences on daily life
in terms of the victim’s ability to work, go to school or do domestic work, their
ability to concentrate on daily activities, levels of confidence and feelings of living
in fear, although few respondents answered this question. The effects were larger
for women than for men, and for physical violence than for other types of violence.
Domestic violence was found to have adverse qualitative effects on child
development, educational opportunities among children and on youth deviant
behaviour, as a result of broken marriages and the resulting economic burden, and
the direct exposure of children to violence within the household – which may result
in those children being more likely to experience or perpetrate domestic violence
later in life.
5.5. Use of and access to victims’ support services and institutions
Just over one in every three women and men who experienced domestic violence
approached a person or formal or informal organisation for help or support. The majority
of respondents who sought help first approached family members or a friend (66.2 per cent
of men and 77.6 per cent of women), whereas 14.6 per cent of men and 9.0 per cent of
women first contacted the police for help. These findings suggest that individuals who
experience domestic violence only resort to external help in extreme circumstances, and
that domestic violence may still be considered largely a private family matter.
Reasons for seeking help included not being able to endure the abuse further (53.3 per cent
of men and 58.3 per cent of women), encouragement by friends and family (28.5 per cent of
men and 24.5 per cent of women), and the extent of injuries sustained (8.0 per cent of men
and 7.5 per cent of women). Reasons for not seeking help included predominantly the fact
that individuals considered the violence as normal or not serious (22.7 per cent of men and
17 per cent of women), and not knowing who to ask (10.7 per cent of men and 16.4 per cent
of women).
Distrust of, and poor experiences with, state authorities and public services contributed to
not seeking help. Factors for these low levels of trust included the costs of the services,
lengthy or ineffective delays in the handling of cases, and corruption. In addition, the
Ghana Police Service had only limited logistical support to encourage the proper
functioning of DOVVSU:
In nine out of ten districts sampled, DOVVSU units or available DOVVSU staff did
not have appropriate separate offices or offer victims privacy when reporting
domestic violence. The only exception was Obuasi in the Ashanti region.
Police personnel had not received adequate specialised training to provide specific
help to the victims of violence.
Observations revealed that police personnel were motivated and professional
despite workloads and lack of resources.
232
In addition to underlying deficiencies in the justice and victims’ support systems,
preferences for traditional over formal legal ways of addressing domestic violence were
also explained by structural inequalities. Notably, poorer victims and those worried about
personal social stigmatisation and reputational loss for the whole family preferred not to
access formal justice processes and resorted to informal processes of mediation. At times,
this was imposed by the social and economic needs of the family or community, and by
prevalent gender norms.
There is room for improvement in terms of raising awareness about the existence of the
Domestic Violence Act and DOVVSU units among Ghana’s general population: 75 per cent
of respondents in urban areas and 61 per cent of respondents in rural areas knew there is
a law against domestic violence in Ghana. There was also a large knowledge variation
across regions, with 10.7 and 12.1 per cent of respondents in the Upper and Northern
regions, respectively, knowing about the law. These were also the regions where the fewest
respondents knew whether there was a DOVVSU unit at their nearest police station.
Satisfaction with the support provided by DOVVSU was low in comparison with other
means of support, but few people answered this question in the survey. Reasons for the
lower levels of satisfaction with the police and DOVVSU mostly related to the direct and
indirect costs of services, including legal services.
233
6. Way forward and policy recommendations
The key purpose of this study was to collect and analyse new evidence about the incidence
levels, determinants and consequences of and attitudes towards domestic violence in
Ghana in 2015 that can be used by the Government of Ghana, CSOs and development
partners to assess current patterns of domestic violence across the country, and to advance
legal, policy and programmatic interventions. To this purpose, this section reflects on how
existing policies and interventions, including the National Domestic Violence Policy and
Plan of Action to Implement the Domestic Violence Act, 2007 (Act 732) (2009–2019) could
be reviewed to better support victims of domestic violence and prevent domestic violence
in Ghana.
The study found that domestic violence directly affects 27.7 per cent of women and 20 per
cent of men in Ghana. In 2015, these women and men experienced at least one form of
violence within domestic relations – either social, physical, sexual, psychological or
economic violence. The causes underlying these levels of domestic violence are complex.
The study found that no single factor or process explained domestic violence victimisation
or perpetration – these varied considerably across the different types of domestic violence
and depended on (i) personal and socio-economic characteristics of victims or perpetrators;
(ii) specific relationship dynamics within families; (iii) norms and structures that operate at
the level of the household, the community and the wider society; and (iv) national- and
global-level factors that shape prevailing gender and social norms, and access to resources
and opportunities by different individuals and social groups.
To address these complex, intersecting factors, policy responses must operate at several
levels and in several sectors. This section discusses which potential interventions may work
– and how existing interventions and programmes may be strengthened – to reduce the
incidence and impact of domestic violence in Ghana. The section is organised around the
main sections of the analysis, and discusses potential recommendations with respect to the
incidence of domestic violence (subsection 6.1), attitudes towards domestic violence
(subsection 6.2), determinants of domestic violence (subsection 6.3), consequences of
domestic violence (subsection 6.4) and use of and access to support services and institutions
(subsection 6.5).
As with other studies, this study uncovered intriguing findings that could not easily be
explained within the scope and time frame of the project, and generated several additional
questions. This section concludes, therefore, with recommendations for future research
work on domestic violence in Ghana. These recommendations focus on four important
areas that were outside the scope of this study: (i) the incidence, causes and consequences
of non-domestic forms of violence, which were found to affect considerable numbers of
women and men in Ghana; (ii) the analysis of causal effects with respect to the causes and
determinants of domestic violence; (iii) attitudes towards and the incidence, causes and
consequences of sexual violence, particularly among men; and (iv) attitudes towards and
234
the incidence, causes and consequences of domestic violence among children, which was
uncovered as an area of concern for future research and policy work.
6.1. Recommendations with respect to the incidence of domestic violence in
Ghana
1. Amendment of the Domestic Violence Act, 2007 (Act 732) to include social abuse as a
specific dimension of domestic violence in the law. The study has shown that social
violence is the second most predominant form of domestic violence against men and
women. It is therefore recommended that the Domestic Violence Act, 2007 (Act 732) is
amended to specifically include this category of abuse within the categories of prohibited
forms of domestic violence in Section 1, in addition to covering physical abuse, sexual
abuse, economic abuse and emotional abuse, as well as harassment. Currently under
Section 1(c) of the Domestic Violence Act it is indicated that domestic violence includes
“harassment including sexual harassment and intimidation by inducing fear in another
person”. In Section 42 of the Domestic Violence Act ‘harassment’ is defined as including
acts of intimidation, threatening acts and some aspects of controlling behaviour. However,
not all the forms of social violence found in the course of the study are set out or obvious
in this definition. It is, therefore, important to specifically include social violence in the
definition of violence in Section 1. The working definition of social violence for the study
was “acts of controlling behaviour experienced by men and women, such as preventing
someone from seeing friends or family of birth; stopping someone from leaving the house;
insisting to know where someone is at all times; stalking; spreading false information,
videos or photos without permission; or forcing women to have an abortion”. This
definition could be adopted in any further amendments of the law.
2. Extend gender-sensitive legislation and awareness and advocacy programmes to bring
attention to non-physical forms of domestic violence. The study has shown that people’s
lived experiences of domestic violence are sometimes at odds with the way in which
domestic violence is conceptualised by policymakers and researchers. This may undermine
effective, tailored policy responses at a practical level. For example, many acts defined as
psychological or economic violence in the Domestic Violence Act were not perceived as
violence by the respondents. These perceptions may result in a mismatch between
legislation and practice, whereby victims of these forms of violence do not feel they ought
to seek support, leading to fewer meaningful policy responses to less overt and less
physical forms of violence. Addressing these issues will involve taking a more
comprehensive approach to gender-responsive legislation in Ghana through gender-
sensitive legislative frameworks that may enable state institutions to develop responsive
policies and transformative protocols to challenge all forms of (physical and non-physical)
domestic violence. These measures should be combined with awareness-raising
campaigns targeting the public, media, police, health, legislators and other decision-
makers and stakeholders about the importance of non-physical forms of domestic violence.
235
3. Improve support available to victims of psychological domestic violence. The study
found that psychological violence was the most common type of violence experienced by
men (affecting 7.9 per cent of men) and the third most common form of domestic violence
experienced by women in Ghana (affecting 9.3 per cent of women). Psychological violence
was also found to be associated with a high risk of women suffering from serious mental
illness. At the same time, psychological violence was generally perceived as a means of
defence and a coping strategy – for women in particular – to ward off physical violence or
to claim some autonomy. These findings required a multi-component approach including
awareness-raising campaigns about the importance and prevalence of psychological
violence, further recognition in support services about this form of violence, strengthening
the skills of mental health professionals to identify and offer protection to those at risk of
this form of abuse (such as psychotherapy and counselling services), and further
cooperation between mental health professionals, NGOs and police units involved in
domestic violence prevention.
4. Further attention is needed to address the widespread prevalence of economic violence.
Economic violence was defined as the denial of household expenses money (chop money)
even if enough financial means are available; unsolicited taking of money; control of
belongings and spending decisions; damage to or destruction of someone’s property;
denial of the right to work; forcing someone to work against their will; or denial of food
and other basic needs. Economic violence was the most prevalent form of domestic
violence against women (affecting 12.8 per cent of women) and the third most common
form of domestic violence against men (affecting 7.3 per cent of men). Economic violence
was particularly prevalent among married women, and less wealthy men and women
living in rural areas, and was found to have severe consequences for the physical and
mental health status of women. These findings call for a stronger coordination of the laws
and policies relating to abuse of individual property rights within domestic relations, and
to economic maintenance provision for married, separated or divorced individuals. There
are provisions on property rights of spouses in Article 22 of the 1992 Constitution and the
Matrimonial Causes Act, 1971 (Act 367), and, apart from the Domestic Violence Act, there
are provisions on economic violence in the Criminal Offences Act, 1960 (Act 29). There will
also be a need to better coordinate domestic violence support services with ongoing
interventions aimed at improving the economic conditions of women (such as
microfinance and cash transfer programmes). Better coordination may ensure a reduction
of the economic dependence of women on men – a factor found to be central to explaining
the prevalence and consequences of domestic economic violence.
5. Extend education, awareness and advocacy programmes to bring attention to domestic
violence against men. The study showed that, although women are more at risk of
experiencing domestic violence, a substantial number of men are also at risk of violence
within domestic relations. Addressing this issue will require bringing attention to forms of
domestic violence experienced by men, and targeting awareness-raising campaigns
towards the value of understanding how men also struggle with harmful masculinities
that might be imposed on them by other men and women. There is also a need to shift
236
social norms about the acceptance of different forms of violence among women – and men.
Further training of police and other support services may also be required, as men find it
even more difficult than women to report cases of domestic violence (when the response
may be disbelief, or ‘giggling’ as some of the qualitative transcripts show).
6. Further attention is needed with respect to child protection. Some forms of physical
violence against children are sometimes perceived as a form of parenting. Non-biological
children can be at particular risk of domestic violence, particularly when the mother is in
a vulnerable economic position. This calls for a policy decision to ensure protection for
children, especially vulnerable non-biological children. It may also call for health and
education workers who engage with children most frequently to receive specialised
awareness training on indicators of domestic violence and the ways available to ensure
that children are protected. Interventions aimed at schools, teachers, children and parents
have reduced violence against children in other countries (see Fulu et al., 2014). Examples
of such interventions include work in schools and with teachers to raise their awareness
about violence-affected children, and improve their skills to behave in non-violent ways;
work with schools and children through curriculum-based awareness-raising
interventions; and work with parents to build skills that involve non-violent means of
parenting. This is, however, an area not explored in detail in the study (the study did not
involve children as participants) and requires further research.
6.2. Recommendations with respect to attitudes towards domestic violence in
Ghana
1. Need for new sensitisation programmes about harmful social norms that drive domestic
violence. The study found that attitudes against any form of domestic violence were strong
and widespread, but many of the definitions of domestic violence used in the study (and
in Act 732) were seen by respondents not as violence but, rather, as accepted social norms,
defence mechanisms or ways of resolving disputes. Only physical violence between
partners was consistently perceived as violence. Psychological violence was generally
perceived as a means of defence and a coping strategy – for women in particular. Other
acts defined as domestic violence in the 2007 Domestic Violence Act – such as controlling
behaviour, verbal abuse or men making sexual remarks to women – were sometimes
deemed acceptable. In particular, wife-beating was accepted as a valid punishment against
disobedience and neglect of children, and rape was considered by the majority of
respondents to be the fault of the woman if she wore revealing clothes. There was also
limited discussion about sexual violence in all focus group discussions and individual
interviews, reflecting the fact that many acts defined as domestic sexual violence may
either be viewed as private or not perceived as violence but, rather, as ‘normal’ behaviour.
These findings suggest that advocacy and awareness-raising campaigns (perhaps through
the use of the media) still have a strong role to play in working to change these widely
accepted norms and behaviours that justify the violence associated with them.
237
2. Need for new sensitisation programmes to reduce acceptance of domestic violence,
particularly among women and girls. The study found that women were on average more
likely than men to find some forms of domestic violence more acceptable. These included
wife-beating, appropriateness of sexual remarks against women, and rape under some
circumstances (if women wore revealing clothes). These findings reflect two important
facts that shape the prevalence of domestic violence in Ghana. First, women who are
victims of domestic violence may have internalised the acceptability of domestic violence
once they are in abusive domestic relations. Second, even when not victims, women may
internalise gender norms that hold women responsible for their experiences of gender-
based violence, rather than placing that responsibility on men or on the broader system
that reinforces gender inequalities. The prevalence of harmful gender norms – such as
discouraging women from engaging in paid labour, expectations that men are the
breadwinner, forcing women to obey men or blaming and shaming women for violence
inflicted on them – was found throughout the study as central to explaining the prevalence,
causes and consequences of domestic violence in Ghana. For instance, the regression
analyses conducted in the study showed that women who expressed a higher acceptance
of women’s sexual autonomy were at greater risk of experiencing psychological, social,
physical or economic violence. The research team interpreted this finding as indicating that
women who challenge established gender norms may be at risk of domestic violence
because they may be punished for their non-compliance with established norms of
behaviour. These findings call for multi-component interventions and strong cooperation
between state agencies and civil society to change the prevalence of harmful gender norms
that perpetuate gender inequalities and condone several forms of domestic violence.
Examples of interventions include (i) economic interventions to reduce the economic
dependence of women on men (such as microfinance, cash-for-work, or cash transfer
programmes); (ii) interventions that work with men and boys to reduce the prevalence of
harmful masculinities and harness them to bring about positive change; (iii) interventions
that work with girls and women to improve their confidence and sense of worth (perhaps
also including psychotherapy); (iv) interventions at the level of family relationships (such
as counselling); and (v) group education campaigns and mobilisation activities at the level
of communities (involving men and women, boys and girls) to raise awareness about the
prevalence of these gender norms and find collective ways to reduce or abandon such
norms altogether.
3. Further awareness-raising is needed about physical violence against men and children.
The qualitative data found a close relationship between participants’ knowledge of legal
rights and acts, such as the Domestic Violence Act, and their perception that physical forms
of domestic violence, especially against women, were unacceptable. Physical violence was
generally interpreted as committed by men against women. Discussions about perceptions
of physical violence committed by women against men revealed that for many people this
was ‘funny’ and sometimes ‘unimaginable’. Physical violence against children, while often
perceived as unacceptable by both men and women, was described in some focus group
discussions, particularly in the Northern and Eastern regions, as a form of education. This
calls for more education and awareness-raising campaigns about the fact that physical
238
violence against anyone, including children, is unacceptable. Interventions discussed
above with respect to the prevalence of sexual violence among men and violence against
children would be useful to increase awareness about the prevalence of attitudes that
condone physical violence against men and children. The Northern and Eastern regions in
particular should be targeted for further interventions in this latter area.
4. Need to target further sensitisation programmes against domestic violence in less
economically developed areas. The study has shown that attitudes against domestic
violence were consistently less pronounced in rural and in the less economically developed
northern areas, with individuals in the lowest asset quintiles and with no education
consistently being more likely to condone several forms of domestic violence than other
socio-economic groups. This finding indicates that sensitisation programmes should target
messages in less developed areas and among populations with lower economic status to
enable a change in attitudes and norms that reinforce domestic violence. The findings also
call for increased awareness-raising in rural communities to ensure an elimination of the
acceptance of abusive behaviour in inter-personal relations and within the domestic
setting. This is particularly important given the intergenerational dynamics of domestic
violence found in this study, and the value of stopping the vicious cycle that adversely
affects people who experience domestic violence at a young age.
6.3. Recommendations with respect to the determinants of domestic violence in
Ghana
1. Need to target interventions towards adolescents and young adults. The findings
indicate strong effects of age and intergenerational factors on the incidence of domestic
violence in Ghana. Young women and men (15–19 years old) were substantially more at
risk of domestic violence than other age groups. Exposure to violence in childhood is also
strongly related to the likelihood of being a victim and perpetrator of violence in
adulthood. These findings suggest that Ghana may experience vicious cycles of domestic
violence reinforced by the fact that young people are more exposed to domestic violence
and that experiencing domestic violence at young ages considerably increases the odds of
further exposure to and perpetration of violence. This calls for targeted interventions,
including sensitisation and therapy for individuals who were exposed to violence at a
young age, to stop the cycle of violence, and coordination with the interventions included
above in Section 6.1 with respect to the incidence of violence against children.
2. Further attention is needed on polygamous relationships. Even though the study found
no clear relationship between marital status and domestic violence, it found that women
in a polygamous relationship were 1.9, 3.5 and 4.3 times more likely to experience
psychological, physical or economic violence, respectively, than never-married women.
Men in polygamous relations were also the most likely to experience sexual or social
domestic violence. This finding emphasises the need to strengthen legislative systems
against polygamy, which violates Article 17 of Ghana’s 1992 Constitution, or to ensure that
239
individuals in polygamous relations are protected under legislation against domestic
violence.
3. Reinforcement of laws and practices about economic maintenance. The study found that
increases in domestic violence were associated with (women’s) refusal to have sex. This
refusal was often used as retaliation against other forms of violence, especially economic
violence. In addition, shirking of economic responsibilities, especially by men not
providing chop money, not paying children’s school fees or not paying other bills, was seen
as an important determinant of domestic violence. In addition to provisions against
economic violence in the Domestic Violence Act, the Children’s Act, 1998 (Act 560) has
provisions that can be used to hold parents responsible to provide for their children. It is,
therefore, important to enforce these laws alongside each other. A recommendation is also
made to support the proposal by the MoGCSP to set up a Child Support Unit to hold non-
custodial parents responsible for contributing to the maintenance of their children, as set
out in the Justice for Children Policy. On the issue of the insistence on the provision of
daily chop money by men for the upkeep of homes, there is a need to work on changing
gender norms, as discussed above, that place women as economically dependent on men,
and expect men to be the sole provider for the family. There may also be the need to explore
whether the expectation for men to provide chop money is simply a means by which men
are made to contribute to housekeeping costs. Encouraging community dialogue (between
men and women) that promotes women’s economic participation, and that values
women’s unpaid labour in the household, might also promote a greater awareness of the
important roles that women can play in the household, alongside men.
4. Reinforcement of property laws. Fights over money, assets or property were found to be
prominent determinants of domestic violence across all regions. Economic tensions were
mentioned to lead to multiple forms of domestic violence among adults – including
between partners, siblings and in-laws. There should be a concerted effort to ensure that
laws in place that protect the property rights of spouses on the death of a spouse – such as
the Intestate Succession Act, 1985 (PNDCL 111) – are disseminated and enforced. With
regard to the property rights of spouses, the Property Rights of Spouses Bill, 2013 that seeks
to actualise Article 22 of Ghana’s 1992 Constitution, which is before Parliament, should be
passed into law as a matter of urgency to protect the property rights of married people at
the dissolution of their marriage or the separation of cohabiting people. The police should
be made aware of these laws to ensure that they are enforced alongside the Domestic
Violence Act.
5. Need to continue sensitisation programmes about alcoholism. Alcoholism was
frequently mentioned as one particularly problematic form of harmful behaviour
associated with domestic violence. Some domestic conflicts were shown to be triggered
following excessive alcohol intake. There is a need for policymakers to continue to educate
people about the harms caused by alcoholism and provide support for the treatment of
alcoholism. There is also a need to advocate that the use of alcohol is not an acceptable
240
excuse or justification for acts of domestic violence and will not stop prosecution for
violations committed.
6. Reinforce legislation and sensitisation interventions against early child marriage. Early
child marriages can be considered a form of domestic violence perpetrated by families
against girls with the consent of both parents. The qualitative data showed evidence of this
practice in some northern regions. Even though the Children’s Act prohibits forced and
early marriage, it is a practice that goes on in some parts of Ghana. A recommendation is
made that the MoGCSP continues undertaking research into the issue and formulates
appropriate policies to address it, including prohibiting it as well as including it in the
Domestic Violence Act.
7. Need to look more closely at the association between domestic violence and community
violence. The results of the study showed a strong association between criminal violence
in the community and incidence of domestic violence. Men who reported violence in their
communities were two to three times more likely than others to report any form of
domestic violence. The odds of women experiencing any form of domestic violence in
communities that experienced other forms of criminal violence increased by 77 per cent for
psychological violence, 49 per cent for physical violence and 57 per cent for economic
violence. The effect was even stronger for sexual violence and social violence, for which
the odds of domestic violence were 3.6 times and 1.9 times higher, respectively, for women
living in communities that experienced other forms of violence, in comparison to women
living in other communities. The qualitative data emphasised in addition that in the
northern regions of Ghana chieftaincy disputes can erupt into community-wide violence.
This finding calls for a national policy to address community violence generally, and
violence associated with chieftaincy and other traditional forms of leadership in specific
places, since this may result in increased incidence of domestic violence. This requires
coordination between different parts of the Ghanaian police dealing with different forms
of violence across communities.
6.4. Recommendations with respect to the consequences of domestic violence in
Ghana
1. Strengthen the involvement of medical practitioners in identifying victims of domestic
violence. The study showed that domestic violence had several adverse consequences for
the physical health of women. These included injuries (particularly because of physical
violence), increased blood pressure and general illnesses. This finding suggests the need
for further cooperation and integration of police and medical practitioners in the
identification and reporting of victims of domestic violence, since the Domestic Violence
Act allows medical practitioners to lodge complaints of domestic violence on behalf of
victims. There is also the need for sensitisation programmes to be undertaken with medical
practitioners.
241
2. Strengthen mental health policy in relation to domestic violence. The results showed a
very strong correlation between exposure to domestic violence and mental health,
particularly for women, and between domestic violence and the inability to perform daily
tasks, feeling depressed and being unable to cope on a daily basis. This finding suggests
that stronger linkages should be forged between domestic violence and mental health
policies. It also calls for stronger and more effective inter-agency collaboration to address
the incidence and harmful consequences of domestic violence. This may include targeted
interventions such as sensitisation and therapy for individuals who experienced domestic
violence, including those who were exposed to violence at a young age. It should also
include supporting the use of psychologists under the Domestic Violence Act and making
the cost of accessing their services affordable to victims and perpetrators.
3. Strengthen child protection policy in relation to domestic violence. The study found that
domestic violence adversely affected child development, educational opportunities among
children and youth behaviour. Domestic violence was also found to have negative impacts
on the education and economic potential of children. This finding suggests that the
MoGCSP must intensify its work around child protection. It suggests also the need for
stronger and more effective inter-agency linkages and strategies to address child protection
more effectively across its various dimensions. There is a particular need to incorporate
information about the adverse consequences of domestic violence in education, awareness-
raising and advocacy programmes, and in work with schools and parents, as effects of
domestic violence go beyond the victim (for instance, on children and on family stability)
and are not widely acknowledged.
6.5. Recommendations with respect to the use of and access to victims’ support
services and institutions in Ghana
1. Reinforce sensitisation programmes about the need to report all forms of domestic
violence. The findings indicated that only about one third of women and men who
experienced at least one form of domestic violence in Ghana in 2015 tried to approach a
person or organisation for help or support. Reasons for seeking help include not being able
to endure the abuse further, encouragement by friends and family, and the extent of
injuries sustained. Individuals exposed to domestic violence seemed to only resort to
external help under more extreme circumstances, and the majority of respondents who
sought help approached family members or a friend first. Reasons for not seeking help
included considering the violence ‘normal’, not knowing who to ask, lack of trust, and
feelings of embarrassment and shame. These findings suggest that domestic violence is still
considered largely a private matter. There is a need for more education and sensitisation
to make people aware that domestic violence is an offence, and not a private matter, and
that this applies to all forms of violence under Act 732. People should be encouraged to
report the occurrence of all forms of domestic violence and should receive adequate
support when turning to formal institutions.
242
2. Improve awareness of the 2007 Domestic Violence Act and available services,
particularly in rural and northern areas. Awareness about domestic violence has improved
since previous studies in Ghana. However, 25 per cent of respondents in urban areas and
39 per cent of respondents in rural areas were not aware of the law against domestic
violence in Ghana. Television and radio were found to be good means of transmission of
information in urban areas (48.3 per cent of urban people were made aware of the Act via
television, and 67 per cent via radio). These means were less effective in rural areas (19.3
per cent of rural respondents knew about the Act via television, and 54.6 per cent via radio).
Building further collaboration between DOVVSU personnel and specialised CSOs, public
units, including the Information Services Department and the National Commission for
Civic Education, media and resource persons in local communities may strengthen
institutional responses to domestic violence. There also seems to be a large variation across
regions in terms of awareness about the law. For instance, 66 per cent of respondents overall
knew that there is a law, in contrast to 10.7–12.1 per cent of respondents in the Upper East,
Upper West and Northern regions. In these regions, very few respondents were also aware
of the presence of a DOVVSU unit. Further efforts are needed to improve awareness of
existing legislation and support channels in these areas.
3. Increase the number, spread, awareness and effectiveness of DOVVSU units. The study
has shown that there is a need for more service units. Only a quarter of survey respondents
in rural areas and 36 per cent in urban areas mentioned that the nearest police station had
a DOVVSU unit. More DOVVSU units across the country would also reduce travel burdens
for victims of domestic violence. It is, however, important to note that part of the reason for
the low usage and knowledge of DOVSSU had also to do with people perceiving traditional
mediation as more acceptable, cheaper and less likely to result in social shaming. The
NPPOA has made provisions for coordination by the DV Secretariat to ensure that various
actors, including CSOs, traditional authorities and others involved in addressing domestic
violence issues, play a role at the community level and report to the regional and national
levels, and there is the need for the Domestic Violence Management Board to ensure that
this coordination system is actualised. It is important that formal state institutions
collaborate with more informal community actors in settling disputes, documenting
settlements and referring matters where necessary – and also in ensuring that different
actors do not further reinforce existing social norms and behaviours that restrict action
against and reporting of domestic violence. There is, therefore, an urgent need to ensure
that the implementing regulations (legislative instruments) that will roll out the Domestic
Violence Act are enacted as soon as possible to implement the Act further, including how
to apply for various protection orders, qualifications required for providing psychological
services, minimum standards for setting up shelters and developing training manuals, and
how to access the Domestic Violence Victim Support Fund.
4. Increase the capacity of service providers to ensure privacy in units and provide
additional support services, such as counselling and shelters. People who experience
domestic violence may have immediate needs beyond medical care, such as access to
therapeutic services, accommodation, legal aid and economic opportunities. Psychological
243
assistance is also often needed by victims of domestic violence who struggle to address
feelings of shame and guilt. The Domestic Violence Act makes provision for these, and the
regulations of the Act are further expected to elaborate on how to access these services. It
is important that the regulations make clear the need for helpline support and information,
shelter/refuge and safe accommodation, short- and long-term psychological counselling,
legal advice, advocacy and outreach services.
5. Build the response capacity of specialised personnel. There is a need to update and
improve the training of police officers to support women and men, boys and girls, as
recommended in the NPPOA. The study found that the staff involved in domestic violence
support services were highly motivated but lacking in wider institutional support.
Additional training and support should enable them to better support the needs of victims
of domestic violence. Consistent training should be encouraged and made part of the short-
and long-term budget of the MoGCSP. Training should also include awareness about
current research findings on the prevalence and gendered dimension of violence globally,
regionally and locally in Ghana. This will encourage the design of appropriate tools to help
specialised support units detect domestic violence at the first contact with women and men
who ask for help, as either victims or perpetrators. It would also help support units to
approach domestic violence respectfully and sensitively.
6. Improve trust in public services that support victims of domestic violence. Distrust of
health care, police and legal services poses a challenge to support services across all regions
of the country. While state institutions may be providing resources to effectively implement
the Domestic Violence Act, they need to engage more directly with how resources are
allocated at a local level and improve the public’s trust in the way in which domestic
violence cases are handled and processed. At a practical level, a free telephone helpline for
people to call in and report cases might be useful. The government could also consider
setting up incentive programmes to encourage police and health care workers to report
cases of domestic violence and follow appropriate procedures. The enactment of the
regulations of the Domestic Violence Act will also strengthen the work of public services
that support victims of domestic violence.
7. Improve inter-agency collaboration. The study observed that district hospitals do not
have specialised services, such as emergency response to individuals who experience
domestic violence, or priority responses for victims of abuse – unless victims are brought
in as normal emergency cases. The MoGCSP should facilitate the development of inter-
agency collaboration between service providers to address this gap so that there is better
collaboration between medical facilities, the police, the judiciary and other institutions
mentioned in the NPPOA. The MoGCSP should also collaborate with the Ministry of Health
to ensure that special spaces are created in district and regional hospitals for victims of
domestic violence.
244
6.6. Recommendations for future research on domestic violence in Ghana
1. Explore further the incidence, determinants and consequences of non-domestic forms of
domestic violence in Ghana. The main aim of this study was to document and analyse the
underlying attitudes towards and incidence, norms, determinants and consequences of
domestic violence in Ghana – i.e. violence that takes place within domestic relations as
defined by the 2007 Domestic Violence Act. The GFLHS 2015 collected, however, a
substantial amount of information on social, physical, sexual, psychological and economic
violence that takes place in non-domestic settings, such as the workplace and the
community. The study found that a large percentage of women and men experienced non-
domestic forms of violence. This is an interesting finding that deserves further exploration
for two main reasons. First, the large incidence of forms of gender-based violence outside
domestic relations suggests that harmful gender norms that condone the use of domestic
violence may also be highly prevalent in the wider society. Therefore, policies that address
violence within domestic relations may miss out vulnerable men and women who continue
to experience forms of violence outside domestic relations. Second, the persistence of
gender-based violence outside domestic settings suggests the widespread acceptance of
gender norms that may contribute to the entrenchment and acceptance of forms of
domestic violence as ‘normal’. The findings of this study suggest an urgent need to explore
further the incidence, determinants and consequences of gender-based violence outside
domestic settings in Ghana – and the data collected in this study could provide the baseline
for such future research.
2. Further research on establishing causality in the determinants and consequences of
domestic violence in Ghana. The multivariate regressions discussed in Section 4 of this
report allowed this study to provide important information about the determinants and
consequences of domestic violence in Ghana that took into consideration potential
correlations between individual variables. This is important because policy action based
on simple cross-tabulations may be misleading, as simple correlations may miss important
interactions between the different complex factors associated with the incidence and
consequences of domestic violence among some groups of people or some regions.
However, the multivariate regressions analysed in this study are only a small step towards
identifying precisely the determinants and consequences of domestic violence. The
multivariate regressions produced in the study provided statistical relationships that take
into account complex interactions between variables but do not allow the identification of
precise causal relations. Establishing the direction of causality among the various factors
that shape domestic violence and among the consequences of domestic violence requires
the use of sophisticated econometric methods that were outside the scope of this study due
to budget and time restrictions.
One example regards findings about the role of education as a determinant of domestic
violence. The study found that the risks of social or sexual violence were higher for men
with higher education levels. This interesting result could be due to a variety of causes. For
instance, it is possible that the risk of domestic violence increases for men with higher
245
education levels because they may have more confidence to report domestic violence. In
this case, domestic violence is higher among this group not due to education per se but
due to other social factors that jointly explain education levels and reporting domestic
violence (for instance, more self-confidence). The analysis provided in this report is not
sufficient to establish the causal reasons that explain the overall finding.
Another example is the relationship reported in the study between violence at the
community level and incidence of domestic violence. The analysis conducted in the study
revealed a strong and statistically significant relationship between these forms of violence
but was not able to identify the direction of causality. It is possible that communities that
experience high levels of community violence may also experience high levels of domestic
violence because violence becomes an accepted norm of behaviour. However, it is also
possible that experiencing domestic violence may predispose individuals to other forms of
violence. These two examples illustrate two statistical challenges – called, respectively,
omitted variable biases and reverse causality – that can only be solved with more
sophisticated quasi-experimental techniques. Efforts should be made to advance causal
research on the determinants and consequences of domestic violence, as this would
uncover more precise entry points for interventions aimed at preventing and remedying
domestic violence and their rigorous evaluation. The datasets collected in this study can
be used for this purpose in future research.
3. Further research on the incidence, causes and consequences of and attitudes towards
sexual violence, particularly against men. This study found that domestic sexual violence
exhibited patterns that were different from all other forms of domestic violence. Notably,
it was more prominent among employed, better-off, less economically dependent
individuals living in urban areas, particularly men. Domestic sexual violence against men
– as well as overall sexual violence over their lifetime and in the 12 months prior to the
GFLHS 2015 – was more common among men in the top asset quintile and with higher
education than among men with lower levels of assets and education. The association for
women was less clear, but there was evidence that better educated women are more likely
to have experienced overall sexual violence (within and outside domestic relations) over
their lifetime and in the 12 months prior to the survey than women with lower levels of
education. Furthermore, while most acts of sexual violence against women were
predominantly perpetrated within domestic relations, most acts of sexual violence against
men were predominantly perpetrated outside domestic relations. It was difficult to
provide a clear explanation for these results without the use of more sophisticated
econometric techniques.
The research team postulated that one likely explanation for these results may be that these
individuals are more likely to have reported these forms of violence, which remain hidden
across most of Ghana. The results discussed in subsection 4.2.3 on attitudes towards
domestic sexual violence provided some evidence that may support this explanation. First,
the study found that there was very limited discussion about sexual violence in the
qualitative data. Second, most survey respondents showed high levels of acceptance of
246
behaviours that the study defined as sexual violence. Taken together, these findings
suggest that sexual violence may be understood largely as something that takes place
outside domestic relations. Perceptions of what may constitute sexual violence within
domestic relations seemed very limited – or hidden. However, much more research is
needed to better understand: (i) the true incidence of domestic sexual violence; and (ii) the
reasons why sexual violence, both domestic and outside domestic relations, was more
prominent against wealthier, more educated men. The research team was also not able to
relate the incidence of this violence to the place where it may take place – for instance, their
workplace, places where they socialise or within the community. There is a need for future
research that makes use of the GFLHS 2015 data to analyse in more depth, using more
sophisticated statistical techniques, the reasons underlying these findings, and also future
studies that analyse in more detail the incidence, causes and consequences of domestic
violence – but also the places where it takes place and the behaviours that may shape its
incidence among particular population groups.
4. Further research on the exposure of children to domestic violence. This study focused on
the analysis of the incidence, determinants and consequences of domestic violence in
Ghana among people aged between 15 and 60 years old. However, the findings reported
in the study – particularly in the qualitative data – highlight the exposure of younger
children to domestic violence. This exposure takes the form of direct victimisation, when
children are beaten or abused in other ways as part of parental disciplining, or indirect
exposure to violence committed against others within their homes. The quantitative data
has shown, in addition, that the adults surveyed in the GFLHS 2015 are more likely to
experience domestic violence if they were exposed directly or indirectly to domestic
violence when they were children. These findings suggest that some families and
communities in Ghana may experience the transmission of norms and attitudes across time
and generations that reinforce the persistence of domestic violence. Addressing this issue
will require more knowledge about the incidence, determinants and consequences of
domestic violence among children under the age of 15. This research was outside the scope
of this study. However, the findings indicate an urgent need for further research in this
area.
247
References
Abane, H. (2003). “For Better, For Worse: Social Dimensions of Marital Conflict in
Ghana:The Case of Cape Coast”, Gender and Behaviour 1: 34–54.
Abramsky, T., Watts, C.H., Garcia-Moreno, C., Devries, K., Kiss, L., Ellsberg, M.,
Jansen,H.A. and Heise, L. (2011). “What Factors Are Associated with Recent
Intimate Partner Violence? Findings from the WHO Multi-country Study on
Women's Health and Domestic Violence”, BMC Public Health 11: 109.
Addai, I. and Adjei, J. (2014). “Predictors of Self-Appraised Health Status in Sub-Saharan
Africa: The Case of Ghana”, Applied Research Quality Life 9: 233–53.
Adinkrah, M. (2014). “Homicide–Suicide in Ghana Perpetrators, Victims, and Incidence
Characteristics”, International Journal of Offender Therapy and Comparative Criminology
58 (3): 364–87.
Adjetey, F.N.-A. (1994). “Reclaiming the African Woman’s Individuality: The Struggle
Between Women’s Reproductive Autonomy and African Society and Culture”, Am.
UL Rev., 44: 1351.
Adomako Ampofo, A.A. (2008). “Collective Activism: The Domestic Violence Bill
becoming Law in Ghana”, African and Asian Studies 7: 395-421.
Adomako Ampofo, A.A. (1993). “Controlling and Punishing Women: Violence against
Ghanaian Women”, Review of African Political Economy 20 (56): 102–11.
Adu-Gyamfi, E. (2014). “Challenges Undermining Domestic Violence Victims’ Access to
Justice in Mampong Municipality of Ghana”, Journal of Law, Policy and Globalization
27 (0): 75–90.
Africa, A. (2010). “Standpoint:“Murderous Women”? Rethinking Gender and Theories of
Violence”, Feminist Africa 14 Rethinking Gender and Violence.
Aizer, A. (2011). “Poverty, Violence, and Health: The Impact of Domestic Violence during
Pregnancy on Newborn Health”, Journal of Human Resources 46 (3): 518–538.
Aizer, A. (2010). “The Gender Wage Gap and Domestic Violence”, American Economic
Review 1847–1859.
Akpalu, E.A., Ofei-Aboagye, E.O. and Derbyshire, H. (2000). “Gender Mainstreaming:
Emerging Lessons from Ghana”.
Alhassan, E. and Odame, F.S. (2015). “Gender Inequality in Basic Education in the
Northern Region of Ghana: Household and Contextual Factors in Perspectives”,
Ghana Journal of Development Studies 12: 125–141.
248
Amoah, J. (2007). “The World on Her Shoulders: The Rights of the Girl-Child in the Context
of Culture and Identity”, Essex Human Rights Review 4 (2): 1–23.
Amoakohene, M.I. (2004). “Violence Against Women in Ghana: A Look at Women’s
Perceptions and Review of Policy and Social Responses”, Social Science & Medicine
59 (11): 2373–2385.
Armstrong, A.K. (1998). Culture and Choice: Lessons from Survivors of Gender Violence
in Zimbabwe. Chapel Hill, NC: AKA.
Artz, L. (2014). “Fear or Failure: Why Victims of Domestic Violence Retract from the
Criminal Justice Process”, South African Crime Quarterly 37: 3–10.
Atinmo, M. (2001). “Sociocultural Implications of Wife Beating Among the Yoruba in
Ibadan City, Nigeria”. In: Oyekanmi, Felicia (ed.), Men, Women and Violence. Dakar:
Council for the Development of Social Science Research in Africa.
Banda, F. (1995). “Provision of Maintenance for Women and Children in Zimbabwe”,
Cardozo Women's LJ 2 (71).
Barker, G. and Ricardo, C. (2005). Young Men and the Construction of Masculinity in Sub-
Saharan Africa: Implications for HIV/AIDS, Conflict, and Violence. Washington, DC:
World Bank.
Bell, C.C. and Mattis, J. (2000). “The Importance of Cultural Competence in Ministering to
African American Victims of Domestic Violence”, Violence Against Women 6: 515–
532.
Belknap, J. (2014). The Invisible Woman: Gender, Crime, and Justice. Boston, MA: Cengage
Learning.
Bloch, F. and Rao, V. 2002. “Terror as a Bargaining Instrument: A Case Study of Dowry
Violence in Rural India”, American Economic Review 92: 1029–1043.
Bograd, M. (1999). “Strengthening Domestic Violence Theories: Intersections of Race,
Class, Sexual orientation, and Gender”, Journal of Marital and Family Therapy 25: 275–
289.
Bourdieu, P. (2001). “Structure, Habitus, Practice”. In: Erickson, P. and Murphy, I. (eds),
Readings for a History of Anthropological Theory. Peterborough, ON: Broadview Press.
Bobonis, G.J., Gonzalez-Brenes, M. and Castro, R. (2013). “Public Transfers and Domestic
Violence: The Roles of Private Information and Spousal Control”, American Economic
Journal: Economic Policy 5 (1): 179–205.
Bowman, C.G. (2003). “Domestic Violence: Does the African Context Demand a Different
Approach?”, International Journal of Law and Psychiatry 26 (5): 473–491.
249
Bowman, C.G. (2002). “Theories of Domestic Violence in the African Context”, Journal of
Gender, Social Policy & the Law 11 (2).
Bowman, C.G. and Schneider, E.M. (1998). “Feminist Legal Theory, Feminist Lawmaking,
and the Legal Profession”, Fordham L. Rev. 67: 249.
Brickell, K. (2015a). “Clouding the Judgment of Domestic Violence Law: Victim Blaming
by Institutional Stakeholders in Cambodia”, Journal of Interpersonal Violence, June
2015: 1–21.
Brickell, K. (2015b). “Towards Intimate Geographies of Peace? Local Reconciliation of
Domestic Violence in Cambodia”, Transactions of the Institute of British Geographers
40 (3): 321–33.
Bronfenbrenner, U. (1979). The Ecology of Human Development: Experiments by Nature and
Design. Cambridge, MA: Harvard University Press.
Burgess, R. and Campbell, C. (2015). “Creating Social Policy to Support Women's Agency
in Coercive Settings: A Case Study from Uganda”, Global Public Health 1–17.
Butegwa, F. (1994). “Using the African Charter on Human Rights and People’s Rights to
Secure Women's Access to Land in Africa”. In: R.J. Cook (ed.), Human Rights of
Women: National and International Perspectives. Philadelphia, PA: University of
Pennsylvania Press: 495.
Buzawa, E.S. and Buzawa, C.G. (2003). Domestic Violence: The Criminal Justice Response. New
York: Sage.
Campbell, C. (1997). “Migrancy, Masculine Identities and AIDS: The Psychosocial Context
of HIV Transmission on the South African Gold Mines”, Social Science & Medicine
45: 273–281.
Carlson, B.E. (1984). “Causes and Maintenance of Domestic Violence: An Ecological
Analysis”, Social Service Review 58: 569–587.
Carrell, S.E. and Hoekstra, M.L. (2010). “Externalities in the Classroom: How Children
Exposed to Domestic Violence Affect Everyone's Kids”, American Economic Journal:
Applied Economics 2 (1): 211–228.
Cerulli, C., Kothari, C., Dichter, M., Marcus, S., Kim, T.K., Wiley, J. and Rhodes, K.V. (2015).
“Help-Seeking Patterns among Women Experiencing Intimate Partner Violence: Do
They Forgo the Criminal Justice System if their Adjudication Wishes Are Not Met?”,
Violence and Victims 30: 16–31.
Clark, C.J., Silverman, J.G., Shahrouri, M., Everson-Rose, S. and Groce, N. (2010). “The Role
of the Extended Family on Women's Risk of Intimate Partner Violence in Jordan”,
Social Science and Medicine 70: 144–151.
250
Coker-Appiah, D. and Cusack, K. (1999). Breaking the Silence and Challenging the Myths of
Violence Against Women and Children in Ghana: Report of a National Study on Violence.
Accra-North, Ghana: Gender Studies & Human Rights Documentation Centre.
Cook, R.J. (2011). Human Rights of Women: National and International Perspectives.
Philadelphia, PA: University of Pennsylvania Press.
Cornwall, A., Harrison, E. and Whitehead, A. (2007). “Gender Myths and Feminist Fables:
The Struggle for Interpretive Power in Gender and Development”, Development and
Change 38: 1–20.
Cornwall, A. and White, S. (2000). “Men, Masculinities and Development: Politics, Policies
and Practices”, IDS Bulletin 31, 1–6.
Coughtry, S.E. (2011). “Patriarchy and the Trap of Masculinity: A Post-Colonial Analysis
of Violence against Sexual Minorities in Uganda”. MSW Thesis. Northampton, MA:
Smith College for Social Work.
Cusack, K. and Manuh, T. (2009) (eds.) The Architecture for Violence against Women in Ghana.
Accra-North, Ghana: Gender Studies & Human Rights Documentation Centre.
http://gendercentreghana.org/?p=632.
Daniel, W.W. (1999). Biostatistics: a foundation for analysis in the health sciences. Hoboken, NJ:
Wiley.
Deaton, A. (1997). The Analysis of Household Surveys: A Microeconometric Approach to
Development Policy. Baltimore, MD: Johns Hopkins University Press.
Dobash, R.E. and Dobash, R. (1979). Violence against Wives. New York: Free Press.
Dobash, R.E. and Dobash, R. (1977). “Love, Honour and Obey: Institutional Ideologies and
the Struggle for Battered Women”, Crime, Law and Social Change 1 (4): 403–415.
Doughty, K.C. (2014). “‘Our Goal Is Not to Punish but to Reconcile’: Mediation in
Postgenocide Rwanda”, American Anthropologist 116: 780–794.
Duvvury, N., Carney, P. and Minh, N. (2012). Estimating the Cost of Domestic Violence in Viet
Nam. New York: UN Women.
Dyson, T. and Moore, M. 1983. “Kinship Structure, Female Autonomy and Demographic
Behavior in India”, Population and Development Review 9: 35–60.
Edström, J., Das, A. and Dolan, C. (2014a). “Introduction: Undressing Patriarchy and
Masculinities to Re-politicise Gender”, IDS Bulletin 45: 1–10.
Edström, J., Izugbara, C., Nesbitt-Ahmed, Z., Otieno, P.E., Granvik, M. and Matindi, S.
(2014b). “Men in Collective Action on SGBV in Kenya: A Case Study”, IDS Evidence
Report 70. Brighton, UK: IDS.
251
Ellsberg, M. and Heise, L. (2005). Researching Violence Against Women: A Practical Guide for
Researchers and Activists. Geneva: World Health Organization, PATH.
Ellsberg, M., Jansen, H.A., Heise, L., Watts, C.H. and Garcia-Moreno, C. (2008). “Intimate
Partner Violence and Women's Physical and Mental Health in the WHO Multi-
country Study on Women’s Health and Domestic Violence: An Observational
Study”, Lancet 371: 1165–1172.
Emmel, N. (2008). Participatory Mapping: An innovative sociological method (Toolkit No. 3).
Southampton, UK: ESRC National Centre for Research Methods.
Eswaran, M. and Malthotra, N. (2011). “Domestic Violence and Women's Autonomy in
Developing Countries: Theory and Evidence”, Canadian Journal of Economics 44 (4):
1222–1263.
Eves, R. (2010). “Masculinity Matters: Men, Gender-Based Violence and the AIDS
Epidemic in Papua New Guinea”, Civic Insecurity: Law, Order and HIV in Papua New
Guinea, 1.
Ewelukwa, U. (2002). “Post-Colonialism, Gender, Customary Injustice: Widows in African
Societies”, Human Rights Quarterly 24: 424–486.
Freedman, J. and Jacobson, S. (2012). “Masculinities, Gender, and Violence”, Engaging Men
in the Fight Against Gender Violence: Case Studies from Africa, 1.
Freeman, M.D. (1977). “Le Vice Anglais? Wife-Battering in English and American Law”,
Family Law Quarterly: 199–251.
Friedman, L. and Couper, S. (1987). The Cost of Domestic Violence: A Preliminary Investigation
of the Financial Cost of Domestic Violence. New York, Victims Services Agency.
Fulu, E., Kerr-Wilson, A., Lang, J., Gibbs, A., Jacobson, J., Jewkes, R., Mazars, C.,
Shauerhammer, V. and Watts, C. 2014. “What Works to Prevent Violence Against
Women and Girls: Evidence Review of Interventions to Prevent Violence Against
Women and Men”.
Gallegos, J.V. and Gutierrez, A. (2011). “The Effect of Civil Conflict on Domestic Violence:
The Case of Peru”, unpublished manuscript.
Garcia-Moreno, C., Jansen, H.A., Ellsberg, M., Heise, L. and Watts, C.H. (2006). “Prevalence
of Intimate Partner Violence: Findings from the WHO Multi-country Study on
Women's Health and Domestic Violence”, Lancet 368: 1260–1269.
García-Moreno, C., Jansen, H.A. Ellsberg, M., Heise, L. and Watts, C.H. (2005). WHO Multi-
country Study on Women’s Health and Domestic Violence Against Women: Initial Results
on Prevalence, Health Outcomes and Women’s Responses. Geneva: World Health
Organization.
252
García-Moreno, C., World Health Organization, London School of Hygiene and Tropical
Medicine and South African Medical Research Council (2013). Global and Regional
Estimates of Violence Against Women: Prevalence and Health Effects of Intimate Partner
Violence and Non-Partner Sexual Violence. Geneva: World Health Organization.
Ghana Statistical Service (GSS) (2013). 2010 Population and Housing Census: National
Analytical Report. Accra: Ghana Statistical Service.
Ghana Statistical Service, Ghana Health Service and ICF Macro (2009). Ghana Demographic
and Health Survey 2008 (DHS Final Reports No. FR221). Accra: Ghana Statistical
Service and Ghana Health Service, and Calverton, MD: ICF Macro.
Geisler, G. (1995). “Troubled Sisterhood: Women and Politics in Southern Africa: Case
Studies from Zambia, Zimbabwe and Botswana”, African Affairs 545–578.
Glaser, B.G. and Strauss, A. (1998). “Grounded Theory”, Strategien qualitativer Forschung.
Bern: Huber.
Goode, W.J. (1971). “Force and Violence in the Family”, Journal of Marriage and the Family
33: 624–636.
Gosselin, D.K. (2000). Heavy hands: An introduction to the crimes of domestic violence. Upper
Saddle River, NJ: Prentice Hall.
Gupte, J., Justino, P. and Tranchant, J.-P. (2014). “Households amid Urban Riots: The
Economic Consequences of Civil Violence in India”, Journal of Conflict Resolution 58
(8): 1445–1473.
Haraway, D. (1988). “Situated Knowledges: The Science Question in Feminism and the
Privilege of Partial Perspective”, Feminist Studies 575–599.
Harrison, A., O’Sullivan, L., Hoffman, S., Dolezal, C. and Morrell, R. (2006). “Gender Role
and Relationship Norms among Young Adults in South Africa: Measuring the
Context of Masculinity and HIV Risk”, Journal of Urban Health 83: 709–722.
Heath, R. (2014). “Women’s Access to Labor Market Oportunities, Control of Household
Resources, and Domestic Violence: Evidence from Bangladesh”, World Development
57 (C): 32–46.
Heise, L.L. (1998). “Violence against Women An Integrated, Ecological Framework”,
Violence Against Women 4(3): 262–290.
Heise, L.L. and Garcia-Moreno, C. (2002). “Violence by Intimate Partners”. In: E.G. King,
L.L. Dahlberg, J.A. Mercy, A.B. Zwi and R. Lozano (eds.), World Report on Violence
and Health. Geneva: World Health Organization.
Hidrobo, M. and Fernald, L. (2013). “Cash Transfers and Domestic Violence”, Journal of
Health Economics 32: 304–319.
253
Hindin, M.J., Kishor, S. and Ansara, D.L. (2008). “Intimate Partner Violence Among
Couples in 10 DHS Countries: Predictors And Health Outcomes”’, DHS Analytical
Studies No. 18. Calverton, MD: Macro International Inc.
Hodgson, D.L. (2002). “Women's Rights as Human Rights: Women in Law and
Development in Africa (WILDAF)”, Africa Today 49: 3–26.
Hoyle, C. (1998). Negotiating Domestic Violence: Police, Criminal Justice, and Victims. Oxford,
UK: Oxford University Press.
Hunnicutt, G. (2009). “Varieties of Patriarchy and Violence against Women: 'Resurrecting
Patriarchy' as a Theoretical Tool”, Violence against Women 15, (5): 553–573.
IDS, Promundo and Sonke Gender Justice (2015). “Engendering Men: A Collaborative
Review of Evidence on Men and Boys in Social Change and Gender Equality”,
EMERGE Evidence Review Report. Brighton, UK: IDS.
Jewkes, R. (2002). “Intimate Partner Violence: Causes and Prevention”, Lancet 359 (9315):
1423–1429.
Jewkes, R., Levin, J. and Penn-Kekana, L. (2002). “Risk Factors for Domestic Violence:
Findings from a South African Cross-Sectional Study”, Social Science & Medicine 55
(9): 1603–1617.
Jones, A. (2006). “Straight as a Rule Heteronormativity, Gendercide, and the
Noncombatant Male”, Men and Masculinities 8: 451–469.
Jones, R.B. (2011). “Development of Feminist Postcolonial Theory”, Postcolonial
Representations of Women. Dordrecht: Springer Netherlands.
Justino, P., Leone, M. and Salardi, P. (2015). “Does War Empower Women? Evidence from
Timor Leste”, IDS Evidence Report 121. Brighton, UK: IDS.
Justino, P., Cardona, I., Mitchell, B. and Muller, C. (2012). “The Role of Women in Post-
Conflict Economic Recovery”. New York: UN Women.
Kelly, K.A. (2003). Domestic violence and the politics of privacy. Ithaca, NY: Cornell University
Press.
Kimuna, S.R. and Djamba, Y.K. (2008). “Gender Based Violence: Correlates of Physical and
Sexual Wife Abuse in Kenya”, Journal of Family Violence 23: 333–342.
Kishor, S. and Bradley, S.E.K. (2012). “Women’s and Men’s Experience of Spousal Violence
in Two African Countries: Does Gender Matter?”, DHS Analytical Studies No. 27.
Calverton, MD: ICF International.
Koenen, K., Moffitt, T., Caspi, A., Taylor, A. and Purcell, S. (2003). “Domestic Violence is
Associated with Environmental Suppression of IQ in Young Children”, Development
and Psychopathology 15 (2): 297–311.
254
Koenig, M.A., Lutalo, T., Zhao, F., Nalugoda, F., Wabwire-Mangen, F., Kiwanuka, N.,
Wagman, J., Serwadda, D., Wawer, M. and Gray, R. (2003). “Domestic Violence in
Rural Uganda: Evidence from a Community-Based Study”, Bulletin of the World
Health Organization 81: 53–60.
La Mattina, G. (2013). Armed Conflict and Domestic Violence: Evidence from Rwanda. Technical
Report. Boston, MA: Boston University.
Little, J. and Jones, O. (2000). “Masculinity, Gender, and Rural Policy”, Rural Sociology 65:
621.
Lloyd, S. (1997). “The Effects of Domestic Violence on Women's Employment”, Law &
Politics 19 (2): 139–167.
Lugo, B.L. (2015). An Exploratory Study on Mental Health Effects of Therapist Minimization of
Domestic Violence Victims’ Experiences. San Diego, CA: Alliant International
University.
Maguigan, H. (2002). “Wading into Professor Schneider’s Murky Middle Ground between
Acceptance and Rejection of Criminal Justice Responses to Domestic Violence”, Am.
UJ Gender Soc. Pol’y & L. 11: 427.
Mann, J.R. and Takyi, B.K. (2009). “Autonomy, Dependence or Culture: Examining the
Impact of Resources and Socio-Cultural Processes on Attitudes Towards Intimate
Partner Violence in Ghana, Africa”, Journal of Family Violence 24 (5): 323–335.
Mann, R.M. (2000). Who Owns Domestic Abuse?: The Local Politics of a Social Problem. Toronto,
ON: University of Toronto Press.
Manuh,T. (2007). “The Passage of Domestic Violence Legislation in Ghana”,
http://r4d.dfid.gov.uk/PDF/Outputs/WomenEmp/GhanaDV.pdf.
Martin, D. (1976). Battered Wives. San Francisco, CA: Glide Publications.
McFadden, P. (2000). “The State of Feminism in Africa Today”, News from the Nordic Africa
Institute 2.
Menon, N. and Johnson, M.P. 2007. “Patriarchy and Paternalism in Intimate Partner
Violence: A Study of Domestic Violence in Rural India”. In: K.K. Misra and J.H.
Lowry (eds), Recent Studies on Indian Women: Empirical Work of Social Scientists.
Jaipur: Rawat Publications.
Mills, E., De Paoli, M.M. and Grønningsaeter, A. (2009). “Love in the Time of AIDS: The
Relational Gender Dynamics of Prevention, Testing and Treatment”, CSSR Working
Paper Series 252. Cape Towm, South Africa: Centre for Social Science Research.
255
Mills, E., Nesbitt-Ahmed, Z., Diggins, J. and Mackieu, T.D. (2015a). ‘They Call Me Warrior’:
The Legacy of Conflict and the Struggle to End Sexual and Gender-based Violence in Sierra
Leone. Brighton, UK: IDS.
Mills, E., Saharokh, T., Wheeler, J., Black, G., Cornelius, R. and Van Den Heever, L. (2015b).
Turning the Tide: The Role of Collective Action for Addressing Structural and Gender-based
Violence in South Africa. Brighton, UK: IDS.
Minayo, M.C.D.S. (1994). “Inequality, Violence, and Ecology in Brazil”, Cadernos de Saúde
Pública 10: 241–250.
Mitchell, L. (2011). Service Users’ Perceptions of the Domestic Violence and Victims’ Support
Unit, Ghana Police Service. Accra: Voluntary Service Overseas (VSO) and Domestic
ViolenceSecretariat.
Moosa, Z. (2012). Violence Against Women and Girls in the Post-2015 Framework: Why & How.
London: ActionAid.
Murphy, J.C. (2002). “Engaging with the State: The Growing Reliance on Lawyers and
Judges to Protect Battered Women”, Am. UJ Gender Soc. Pol’y & L. 11: 499.
NETRIGHT (2012). “Ghana NGOs Alternative Report for CSW 57,” Accra-North:
NETRIGHT, Association of African Women in Research & Development
(AAWORD) and UN Women Ghana Country Programme Office.
Njovana, E. and Watts, C. (1996). “Gender Violence in Zimbabwe: A Need for
Collaborative Action”, Reproductive Health Matters 4: 46–55.
Noe, D. and Rieckmann, J. (2013). “Violent Behaviour: The Effect of Civil Conflict on
Domestic Violence in Colombia”, Poverty, Equity and Growth Discussion Paper.
Göttingen, Germany: Courant Research Centre.
Ofei-Aboagye, R.O. (1994a). “Domestic Violence in Ghana: An Initial Step”, Colum. J.
Gender & L., 4: 1.
Ofei-Aboagye, R.O. (1994b). “Initial Questions”, The Public Nature of Private Violence: The
Discovery of Domestic Abuse. New York: Routledge: 260.
Ofei-Aboagye, R.O. (1994c). “Altering the Strands of the Fabric: A Preliminary Look at
Domestic Violence in Ghana”, Signs 19 (4): 924–938.
Panda, P. and Agarwal, B. (2005). “Marital Violence, Human Development and Women's
Property Status in India”, World Development 33: 823–850.
Perilla, J.L., Bakeman, R. and Norris, F.H. (1994). “Culture and Domestic Violence: The
Ecology of Abused Latinas”, Violence and Victims 9: 325–339.
Pollack, R.A. (2004). “An Intergenerational Model of Domestic Violence”, Journal of
Population Economics 17: 311–329.
256
Pool, M.S., Otupiri, E., Owusu-Dabo, E., de Jonge, A. and Agyemang, C. (2014). “Physical
Violence during Pregnancy and Pregnancy Outcomes in Ghana”, BMC Pregnancy
and Childbirth 14 (1): 71.
Ribero, R. and Sanchez, F. (2005). “Determinants, Effects and Costs of Domestic Violence”,
Documento CEDE 2005-38. Bogota, Colombia: Universidad de los Andes.
Richards, P. (2005). “To Fight or to Farm? Agrarian Dimensions of the Mano River Conflicts
(Liberia and Sierra Leone)”, African Affairs 104: 571–590.
Richards, P. (1996). Fighting for the Rain Forest: War, Youth and Resources in Sierra Leone.
Oxford: James Currey.
Rowe, M. (2015). “Community Impact Statements in Domestic Violence Criminal Cases: A
World First”, Legaldate 27: 11.
Schneider, E.M. (2000). Battered Women and Feminist Lawmaking. New Haven, CT: Yale
University Press.
Schneider, E.M. (2008). “Domestic Violence Law Reform in the Twenty-First Century:
Looking Back and Looking Forward”, Fam. LQ 42: 353.
Shanley, M.L. (1993). Feminism, Marriage, and the Law in Victorian England. Princeton, NJ:
Princeton University Press.
Shepard, M. and Pence, E. (1988). “The Effect of Battering in the Employment Status of
Women”, Affilia 3 (2): 55–61.
Sherman, L.W., Schmidt, J.D. and Rogan, D.P. (1992). Policing Domestic Violence: Experiments
and Dilemmas. New York: Free Press.
Sipsma, H., Ofori-Atta, A., Canavan, M., Osei-Akoto, I., Udry, C. and Bradley, E. (2013).
“Poor Mental Health in Ghana: Who Is at Risk?”, BMC Public Health 13 (1): 288.
Sokoloff, N.J. and Dupont, I. (2005). “Domestic Violence at the Intersections of Race, Class,
and Gender Challenges and Contributions to Understanding Violence Against
Marginalized Women in Diverse Communities”, Violence against Women 11: 38–64.
Stockl, H., Devries, K., Rotstein, A., Abrahams, N., Campbell, J., Watts, C. and Garcia-
Moreno, C. (2013). “The Prevalence of Intimate Partner Homicide: A Systematic
Review”, Lancet 382.
Sternberg, K.J., Lamb, M.E., Greenbaum, C., Cicchetti, D., Dawud, S., Cortes, R., Krispin,
O. and Lorey, F. (1993). “Effects of Domestic Violence on Children's Behavior
Problems and Depression”, Developmental Psychology 29 (1): 44–52.
Sulak, T.N., Saxon, T.F. and Fearon, D. (2014). “Applying the Theory of Reasoned Action
to Domestic Violence Reporting Behavior: The Role of Sex and Victimization”,
Journal of Family Violence 29: 165–173.
257
Takyi, B.K. and Mann, J. (2006). “Intimate Partner Violence in Ghana, Africa: The
Perspectives of Men regarding Wife Beating”, International Journal of Sociology of the
Family 32 (1): 61–78.
Tamale, S.R. (1993). “Law Reform and Women's Rights in Uganda”, East African Journal of
Peace and Human Rights 1 (2): 164–94.
Tanchen, H.V., Witte, A.D. and Long, S.K. (1991). “Domestic Violence: A Nonrandom
Affair”, International Economic Review 32: 491–511.
Tenkorang, E.Y., Owusu, A.Y., Yeboah, E.H. and Bannerman, R. (2013). “Factors
Influencing Domestic and Marital Violence against Women in Ghana”, Journal of
Family Violence 28 (8): 771–81.
Thaler, K. (2012). “Norms About Intimate Partner Violence among Urban South Africans:
A Quantitative and Qualitative Vignette Analysis”, CSSR Working Paper No. 302.
Cape Towm, South Africa: Centre for Social Science Research, Social Surveys Unit.
Tsikata, D. and Seini, W. (2004). “Identities, Inequalities and Conflicts in Ghana”, CRISE
Working Paper 5. Oxford: Centre for Research on Inequality, Human Security and
Ethnicity.
Tsikata, P.Y. (2014). “The Historical and Contemporary Representation of Africa in Global
Media Flows: Can the Continent Speak Back for Itself on Its Own Terms?”,
Communication 40: 34-48.
Turshen, M. and Turshen, M. (2000). “Even Strong Backs Need Support: Introductory
Notes”, African Women’s Health 8.
UNITAR (2011). “River of Life Method”, www.click4it.org.
USAID (2006). Addressing Gender-based Violence through USAID’s Health Programs: A Guide
for Health Sector Program Officers. Washington, DC: USAID.
Walker, L.E. (1999). “Psychology and Domestic Violence Around the World”, American
Psychologist 54: 21.
Watts, C., Osam, S. and Win, E. (1995). “The Private is Public: A Study of Violence against
Women in Southern Africa”. Harare: Zimbabwe: Women in Law and Development
in Africa.
Wolfe, D.A., Crooks, C.V., Lee, V., McIntyre-Smith, A., Jaffe, P.G. (2003), “The Effects of
Children's Exposure to Domestic Violence: A Meta-Analysis and Critique”, Clinical
Child and family Psychology Review 6 (3): 171–187.
WHO (2001). “Putting Women First: Ethical and Safety Recommendations for Research on
Domestic Violence against Women”. Geneva: World Health Organization,
http://www.who.int/gender/documents/violence/who_fch_gwh_01.1/en/.
258
Wilson, M. and Daly, M. (1996). “Male Sexual Proprietariness and Violence against Wives”,
Current Directions in Psychological Science 5: 2–7.
Wilson, M. and Daly, M. (1993). “An Evolutionary Psychological Perspective on Male
Sexual Proprietariness and Violence against Wives”, Violence and Victims 8: 271–294.
Yllo, K.A. and Strauss, M.A. (1990). “Patriarchy and Violence against Wives: The Impact of
Structural and Normative Factors”. In: M.A. Strauss and R.J. Gelles (eds), Physical
Violence in American Families. New Brunswick, NJ: Transaction Publishers.
Yodanis, C.L. (2004). “Gender Inequality, Violence against Women, and Fear A Cross-
National Test of the Feminist Theory of Violence Against Women”, Journal of
Interpersonal Violence 19: 655–675.
259
Appendices
260
Appendix A: Incidence of domestic violence worldwide
Prevalence of physical
violence against women (%)
Prevalence of sexual violence
against women (%)
All
perpetrators
By intimate
partner
All
perpetrators
Intimate
partner
Region Country or area Year Life
time
Last
12
mths
Life
time
Last
12
mths
Life
time
Last
12
mths
Life
time
Last
12
mths
Africa Burkina Faso 2010 19.8 9.4 11.1 9.2 .. .. 1.5 1.2
Africa Cabo Verde 2005 21.5 20.3 15.7 .. .. .. 3.6 ..
Africa Cameroon 2011 54.6 27.4 44.8 29.0 29.0 9.8 20.3 11.2
Africa Central African
Republic
2006 35.3 28.7 25.4 22.6 .. .. 11.6 8.3
Africa Chad 2010 .. .. .. 18.4 .. .. .. 12.0
Africa Comoros 2012 14.0 5.8 5.6 4.2 6.0 1.3 1.8 1.3
Africa Côte d'Ivoire 2011-12 35.6 19.9 24.6 22.2 4.5 .. 5.3 4.6
Africa Democratic
Republic of the
Congo
2007 63.7 49.0 56.9 .. 16.0 4.2 35.3 ..
Africa Egypt 2005 47.4 15.7 33.2 18.2 .. .. 6.6 3.9
Africa Equatorial Guinea 2011 62.8 55.8 54.4 40.2 31.9 8.2 17.4 12.5
Africa Ethiopia -
Province
2002 .. .. 48.7 29.0 .. .. 58.6 44.4
Africa Gabon 2012 52.4 22.3 46.2 28.3 20.8 8.3 17.0 11.8
Africa Ghana 2008 36.6 17.2 20.6 18.0 18.8 .. 8.2 5.2
Africa Kenya 2008-09 38.5 24.0 37.0 31.3 20.6 .. 17.2 13.6
Africa Liberia 2007 44.0 28.9 35.0 33.0 17.6 .. 10.8 9.6
Africa Malawi 2010 28.2 14.2 21.7 14.7 25.3 .. 18.9 13.4
Africa Mali 2012-13 38.3 24.7 29.8 20.7 12.6 10.5 13.9 12.1
Africa Morocco 2009-10 35.3 15.2 .. 6.4 22.6 8.7 .. 6.6
Africa Mozambique 2011 33.4 25.0 31.5 25.9 12.3 6.9 7.9 6.9
Africa Namibia – City 2002 .. .. 30.6 15.9 .. .. 16.5 9.1
Africa Nigeria 2013 27.8 11.2 14.4 9.3 7.4 3.3 4.8 3.7
Africa Rwanda 2010 41.2 .. 55.6 32.8 22.3 .. 17.5 13.3
Africa Sao Tome and
Principe
2008-09 33.4 21.1 26.5 .. 13.1 .. 8.3 ..
Africa Sierra Leone 2013 55.5 27.0 44.2 27.2 10.5 4.9 7.3 5.1
Africa South Africa 1998 .. .. 12.5 6.3 .. .. 4.4 ..
Africa Tunisia 2010 31.7 7.3 20.3 7.2 15.7 7.4 14.2 9.0
Africa Uganda 2011 56.1 26.9 42.7 24.9 27.8 16.2 27.3 20.9
Africa United Republic
of Tanzania
2010 38.7 33.0 39.2 33.4 20.3 .. 17.2 13.7
Africa Zambia 2007 46.8 32.5 46.5 39.6 20.2 .. 16.7 16.0
Africa Zimbabwe 2010-11 29.9 18.4 28.8 20.7 27.2 .. 26.0 13.3
Asia Armenia 2008 .. .. 8.9 .. .. .. 3.3 ..
Asia Azerbaijan 2006 13.3 7.7 12.8 9.7 3.7 .. 2.9 2.0
Asia Bangladesh 2011 .. .. 64.6 46.4 .. .. 36.5 24.1
261
Asia Cambodia 2005 22.3 10.3 12.8 7.9 .. .. 2.7 1.7
Asia China 1999-00 .. .. 15.4 .. .. .. .. ..
Asia China, Hong
Kong
2005 12.0 2.0 6.0 1.0 14.0 3.0 5.0 1.0
Asia Cyprus 2014 20.0 5.0 14.0 3.0 5.0 1.0 4.0 1.0
Asia Georgia 2010 .. .. 4.5 1.4 .. .. 1.7 0.5
Asia India 2005-06 33.5 18.9 35.1 21.4 8.5 .. 10.0 7.2
Asia Japan 2010 .. .. 25.9 .. .. .. 14.1 ..
Asia Jordan 2012 34.3 12.6 21.1 11.2 9.2 .. 8.6 6.0
Asia Kyrgyzstan 2012 23.1 13.3 25.1 16.9 3.4 2.1 4.0 2.8
Asia Maldives 2006 .. .. 17.9 5.7 .. .. 6.7 2.0
Asia Mongolia 2008 .. .. .. .. 4.0 .. .. ..
Asia Nepal 2011 21.5 9.3 23.1 10.4 12.3 6.4 14.3 7.7
Asia Pakistan 2012-13 32.2 19.2 26.8 18.0 .. .. .. ..
Asia Philippines 2013 19.6 5.6 12.7 5.3 6.3 2.7 5.3 3.2
Asia Republic of Korea 2013 .. .. .. 7.2 19.5 2.7 .. 5.4
Asia Singapore 2009 6.8 1.0 5.7 0.9 4.2 0.3 1.2 0.1
Asia State of Palestine 2011 .. .. 30.6 23.5 .. .. 14.6 11.8
Asia Tajikistan 2012 18.8 13.0 19.5 14.5 3.7 2.5 4.4 3.3
Asia Thailand -
Province
2002 .. .. 33.8 13.4 .. .. 28.9 15.6
Asia Thailand - City 2005 .. .. 22.9 7.9 .. .. 29.9 17.1
Asia Timor-Leste 2009-10 38.1 29.2 33.5 30.7 3.4 .. 2.9 2.0
Asia Turkey 2014 .. .. 36.0 8.0 .. .. 12.0 5.0
Asia Viet Nam 2010 35.2 .. 31.5 6.4 10.8 .. 9.9 4.2
Europe Albania 2013 .. .. 23.7 14.7 .. .. 7.9 5
Europe Austria 2014 17.0 4.0 12.0 2.0 9.0 2.0 6.0 1.0
Europe Belgium 2014 33.0 10.0 22.0 5.0 13.0 2.0 9.0 1.0
Europe Bulgaria 2014 27.0 7.0 22.0 6.0 12.0 3.0 9.0 3.0
Europe Croatia 2014 19.0 5.0 12.0 3.0 5.0 1.0 3.0 0.0
Europe Czech Republic 2014 30.0 7.0 19.0 4.0 9.0 2.0 7.0 1.0
Europe Denmark 2014 48.0 10.0 29.0 3.0 19.0 2.0 11.0 1.0
Europe Estonia 2014 30.0 4.0 19.0 2.0 13.0 2.0 7.0 1.0
Europe Finland 2014 43.0 9.0 27.0 4.0 17.0 3.0 11.0 1.0
Europe France 2014 42.0 11.0 25.0 5.0 15.0 2.0 9.0 1.0
Europe Germany 2014 33.0 7.0 20.0 3.0 12.0 1.0 8.0 1.0
Europe Greece 2014 24.0 7.0 18.0 5.0 6.0 2.0 5.0 2.0
Europe Hungary 2014 25.0 8.0 19.0 5.0 9.0 3.0 7.0 2.0
Europe Iceland 2008 29.8 2.1 19.7 1.2 24.2 1.6 6.1 0.1
Europe Ireland 2014 24.0 7.0 14.0 3.0 8.0 2.0 6.0 1.0
Europe Italy 2014 25.0 6.0 17.0 5.0 9.0 4.0 7.0 4.0
Europe Latvia 2014 36.0 6.0 31.0 5.0 14.0 2.0 9.0 1.0
Europe Lithuania 2014 30.0 5.0 24.0 4.0 7.0 0.0 4.0 0.0
Europe Luxembourg 2014 36.0 6.0 21.0 3.0 15.0 2.0 9.0 1.0
Europe Malta 2014 19.0 4.0 13.0 3.0 9.0 2.0 6.0 1.0
Europe Netherlands 2014 41.0 9.0 22.0 4.0 18.0 3.0 11.0 2.0
262
Europe Norway 2008 .. .. 13.9 .. .. .. 9.4 1.0
Europe Poland 2014 18.0 4.0 12.0 2.0 5.0 1.0 4.0 1.0
Europe Portugal 2014 23.0 5.0 18.0 4.0 4.0 1.0 3.0 1.0
Europe Republic of
Moldova
2010 .. .. 39.7 8.9 .. .. 18.6 4.1
Europe Romania 2014 29.0 7.0 23.0 6.0 6.0 2.0 5.0 2.0
Europe Serbia - City 2003 .. .. 22.8 3.2 .. .. 6.3 1.1
Europe Slovakia 2014 33.0 9.0 22.0 6.0 10.0 3.0 8.0 2.0
Europe Slovenia 2014 21.0 3.0 12.0 2.0 7.0 1.0 4.0 0.0
Europe Spain 2014 20.0 3.0 12.0 1.0 6.0 1.0 4.0 1.0
Europe Sweden 2014 41.0 8.0 24.0 3.0 18.0 3.0 10.0 2.0
Europe Switzerland 2003 27.0 1.0 9.0 1.0 25.0 1.0 3.0 ..
Europe Ukraine 2007 16.9 8.8 12.7 10.4 5.4 .. 3.3 2.2
Europe United Kingdom 2014 42.0 8.0 28.0 4.0 14.0 2.0 10.0 1.0
LA & Carib. Bolivia 2008 .. .. 23.1 .. .. .. 6.4 ..
LA & Carib. Brazil - City 2000-03 .. .. 27.2 8.3 .. .. 10.1 2.8
LA & Carib. Brazil - Province 2000-03 .. .. 33.8 12.9 .. .. 14.3 5.6
LA & Carib. Colombia 2010 .. .. 37.4 .. .. .. 9.7 ..
LA & Carib. Costa Rica 2003 47.0 11.0 33.0 7.0 41.0 7.0 15.0 3.0
LA & Carib. Dominican
Republic
2007 19.7 9.7 16.1 10.9 10.1 .. 5.5 3.6
LA & Carib. Ecuador 2011 38.0 .. 35.0 .. 25.7 .. 14.5 ..
LA & Carib. El Salvador 2008 .. .. 24.2 6.8 13.4 .. 11.5 3.3
LA & Carib. Guatemala 2008-09 .. .. 24.5 7.8 5.8 .. 12.3 4.8
LA & Carib. Haiti 2012 28.2 10.4 15.6 10.3 13.1 .. 11.1 8.6
LA & Carib. Honduras 2011-12 27.0 .. .. 10.0 4.0 .. .. 3.2
LA & Carib. Jamaica 2008-09 .. .. 17.0 6.5 12.0 .. 7.6 2.8
LA & Carib. Mexico 2011 15.2 6.4 12.3 5.4 38.9 20.8 5.9 2.5
LA & Carib. Nicaragua 2006-07 19.2 .. 27.0 8.0 11.4 .. 13.1 4.4
LA & Carib. Paraguay 2008 .. .. 17.9 6.7 .. .. 5.0 1.7
LA & Carib. Peru 2013 .. .. 35.7 11.5 .. .. 8.4 3.0
LA & Carib. Puerto Rico 1995-96 .. .. .. .. 9.1 .. 5.7 ..
Northern
America
Canada 2009 .. 3.4 .. 1.3 .. 2.0 .. ..
Northern
America
United States of
America
2011 .. .. 31.5 4.0 19.3 1.6 8.8 0.8
Oceania Australia 2012 34.0 4.6 15.6 .. 19.0 1.2 4.9 ..
Oceania Fiji 2010-11 68.5 .. 61.1 19.4 35.6 .. 33.9 14.2
Oceania Kiribati 2008 .. .. 60.0 32.4 .. .. 46.4 33.7
Oceania New Zealand -
City
2002 .. .. 30.2 5.3 .. .. 14.1 2.1
Oceania Samoa 2000 .. .. 40.5 17.9 .. .. 19.5 11.5
Oceania Solomon Islands 2009 .. .. 45.5 .. .. .. 54.7 ..
Oceania Tonga 2009 76.8 .. 33.4 12.5 17.4 .. 16.5 11.0
Oceania Vanuatu 2009 .. .. 51.0 33.0 .. .. 44.0 33.0
263
Appendix B: Timeline of major policies related to domestic and gender-based
violence International Regional National
1948•Universal Declaration of Human Rights
1966•International Covenant on Civil and Political Rights•International Covenant on Social, Economic and Cultural Rights
1981•African Charter on Human and Peoples' Rights
1982•Convention on the Elimination of All Forms of Discrimination Against Women (CEDAW)
1992•Republic of Ghana Constitution - does not include domestic violence or violence against women
but does include equality for women
1993•Declaration on the Elimination of Violence Against Women - comprehensive set of standards in
international law for the protection of women against sexual and gender-based violence
1994•Republic of Ghana Criminal Code Article 69A - makes female genital mutilation a second-degree
felony
1994•African Platform for Action on Women - Dakar Declaration - "violation of women's rights is
violation of human rights" … appointment of a Special Rapporteur on Violence Against Women
1995•Protocol to the African Charter on Human and People's Rights on the Rights of Women in Africa
- includes the need for States to adopt legislation aimed to eradicate violence against women
1995
•Beijing Declaration and Platform for Action - Conference Report from Fourth World Conference on Women - Countries (including Ghana) that have signed up to the Beijing Declaration are requried to give five-yearly reports detailing the action they have taken to improve the situation for women and girls
1999
•African Commission on Human and People's Rights recruits a Special Rapporteur on the Rights of Women with a mandate to servce as a focal point for promotion and protection of the rights of women
2000•UN Resolution 55/68 - Elimination fo all forms of violence, including crimes against women
264
2003•UN Resolution 58/185 - In-depth study on all forms of violence against women
2003•Ghanian Poverty Reduction Strategy (GPRS) - includes the necessity to formulate and enact
Domestic Violence Act. Put together in consultation with the International Monetary Fund
2003
•Ghanaian Strategic Implementation Plan - includes 'mainstreaming of gender issues', decentralisation of 'women's machinery' and monitoring of policies affecting women and children
2004•UN Resolution 18/147 Elimination of Domestic Violence Against Women - specifically
addressing domestic violence as a separate issue for the first time
2004•African Union - Solemn Declaration on Gender Equality and Women's Empowerment - no
specific mention of domestic violence
2006•UN Resolution 61/143 - intensification of efforts to eliminate all forms of violence against
women
2007•African Commission on Human and People's Rights - Resolution 111 - On the right to remedy
and reparation for women and girl victims of sexual violence
2007•Republic of Ghana Domestic Violence Act
2008•UN Resolution 7/24 - Elimination of Violence Against Women
2009•UN Resolution 11/2 - Accelerating efforts to eliminate all forms of violence against women•UN Resolution 12/17 - Elimination of discrimination against women
2010
•GHANAP 1325 - the Ghanaian Action Plan for the implementation of UN Resolution 1325 (women, peace, security). Although not explicitly about domestic violence, the mechanisms that have been establised to protect women from domestic violence will be adopted to protect women in situations of conflict and gender-based violence
2010
•Ghanaian Sector Medium-Term Development Plan (SMTDP)- includes the need to education women regarding their human rights and access to the law - for example, the Domestic Violence Act
2010•Ghanaian Shared Growth and Development Agenda (GSGDA) - the Domestic Violence Act is seen
as the solution to women's inequality
265
Appendix C: Summary of qualitative sample, methods and instruments
Method of data
collection
Sampling Sample characteristics Instruments
FOCUS GROUP
DISCUSSIONS
One urban and one
rural site from one
district in each
region
1 hour /8–12 people
80 FGDs: 8 per region (4 in
rural, 4 in urban sites)
Group 1: Married/living
together men (aged 18–50)
Group 2: Married /living
together women (aged 18–50)
Group 3: Unmarried/youth
(aged 18–25)
Group 4: Mixed (aged 50+)
Semi-structured
questionnaire
OBSERVATIONS
AND IN-DEPTH
INTERVIEWS
Primary public
institutions of
victim support:
police and health in
one urban and one
rural site from one
district in each
region
Interviews with 20 DOVVSU
officers (2 from each chosen
district office in each region)
Observations conducted in
each of the 10 district
DOVVSUs visited where staff
were interviewed
28 health staff interviewed,
including doctors, nurses and
hospital administrators; 10
district hospitals were
observed
Observations
using a checklist
with additional
descriptions
Semi-structured
questionnaire
IN-DEPTH
INTERVIEWS
with OPINION
LEADERS,
ACTIVISTS and
LEGAL EXPERTS
Purposive sample
in all regions
equally divided
across urban and
rural areas
Total of 200 opinion leaders:34
10 legal practitioners (one per
region); 20 activists (2 per
region); 20 media persons (2
per region); 102 community
leaders as described in the
text
Observations
using a checklist
with additional
descriptions
Semi-structured
questionnaire
34 Legal practitioners, media and activists were mainly located at urban sites.
266
Appendix D: Distribution of enumeration areas across regions and survey
weights
Table D1: Distribution of enumeration areas
Region
2010
population
in household
distribution
2010
population
in household
distribution
per cent
Propor-
tionate
allocation
of
selected
EAs
Adjusted
number of
HHs
selected
per EA
Propor-
tionate
allocation
of
selected
EAs
Adjusted
number of
HHs
selected
per EA
Western 2,307,395 9.6 32 480 32 480
Central 2,113,766 8.9 29 435 29 435
Greater Accra 3,888,512 16.3 53 795 53 795
Volta 2,086,567 8.6 29 435 29 435
Eastern 2,574,549 10.7 35 525 35 525
Ashanti 4,671,982 19.4 64 960 64 960
Brong Ahafo 2,265,458 9.4 31 465 31 465
Northern 2,445,061 10.1 34 510 34 510
Upper East 1,034,704 4.2 14 210 14 210
Upper West 688,333 2.8 9 135 9 135
Total 24,076,327 100.0 331 4,950 331 4,950
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
Weighting of data
The survey data were weighted at the data management stage, taking into account the
probability of selecting a primary sampling unit from a region, the probability of selecting
a household from the primary sampling unit, and the probability of selecting a qualified
individual for interviewing. The weight was standardised to be applicable to the data. This
was because the sampling design was not self-weighting: households were selected with
unequal probabilities at the second stage of sampling. As a result, a sampling weight
accounted for differences in the selection probability and in the number of completed
interviews among primary sampling units. The required information for computing the
weights at the level of the primary sampling unit and household selection stage are
explained below.
Computation of weights
The weight is the product of the first- and second-stage probabilities of selection. The first-
stage probability of selection was calculated based on the number of clusters selected in
the stratum (ah), multiplied by household size (Mhi) (among those aged 15–60 years) and
divided by stratum size (∑Mhi). The second-stage probability of selection was computed as
the number of households selected in the clusters i (b) divided by the number of
households listed in cluster i (Mhi*) (i.e. the probability of selecting 15 households aged 15–
267
60 years out of the listed households aged 15–60 years). The reciprocal of the overall
probability of selection is the weighting factor, also known as the design weight.
Let Mhi = Number of 2010 Population Census households in the ith selected
enumeration area in the hth stratum or region
Mhi* = Number of households listed in the ith selected enumeration area in the hth
stratum (urban/rural in the region)
ΣMhi = Total number of households in the ith stratum (i.e. number of households
in
either an urban or rural areas in a region)
ahi = Number of sample enumeration areas allocated to the hth stratum (urban/rural
in the region)
e.g. a11 = 14 for urban area in the Western region
and a12 = 18 for a rural area in the Western region
b = 15 (number of selected households per enumeration area in each stratum)
Then, the first- and second-stage probabilities of selection are:
hi
hihihi
M
MaP
1 and
*2
hi
hiM
bP
Where,
P1 hi is the probability of selecting the ith enumeration area in the hth stratum, and P2 hi is
the probability of selecting a household in the ith enumeration area of the hth stratum. The
overall probability of selection of a household in the ith selected enumeration area of the
hth stratum is the design weight given by:
hihihi PPF 21 *
*
*hi
hi
hi
hihi
M
M
M
baF
The weighting factor (or expansion factor), Whi, for a household in the ith selected
enumeration area in the hth stratum is the reciprocal (inverse) of the overall probability of
selecting that household. That is,
hi
hiF
W1
268
hi
hi
hi
hi
M
M
ba
M **
The number of households successfully interviewed in each enumeration area was used in
the computation. Therefore, the final weight for the sample households in the jth cluster
and in the ith sample PSU in stratum h is given by:
hihi Wb
bW *
''
''
Where:
'b the number of interviews plus the number of no interviews in the sample cluster
''b the total number of interviewed sample households selected in the jth sample PSU
within the ith sample stratum h.
269
Appendix E: Regression tables
Table E1: Determinants of women having experienced domestic violence in the 12
months prior to survey: odds ratios estimated using logistic regressions
Psychological Physical Sexual Social Economic
violence violence violence violence violence
Age category (r:
15–19)
20–24 0.54** 0.50** 0.44* 0.52*** 0.77
(0.13) (0.14) (0.22) (0.12) (0.22)
25–29 0.42*** 0.34*** 0.37** 0.44*** 1.12
(0.11) (0.095) (0.19) (0.10) (0.31)
30–39 0.35*** 0.31*** 0.17*** 0.23*** 0.69
(0.087) (0.095) (0.088) (0.058) (0.19)
40–49 0.21*** 0.12*** 0.033*** 0.15*** 0.40***
(0.068) (0.045) (0.028) (0.045) (0.13)
50–60 0.16*** 0.0086*** 0.020*** 0.11*** 0.27***
(0.054) (0.0090) (0.021) (0.036) (0.088)
Work situation (r:
self-employed)
Employed 0.80 0.76 0.95 1.35 0.73
(0.20) (0.25) (0.39) (0.27) (0.18)
Not working 1.20 1.10 0.81 1.23 1.43**
(0.23) (0.25) (0.25) (0.19) (0.22)
Asset Index 0.95** 0.96 1.01 0.98 0.95**
(0.022) (0.025) (0.011) (0.011) (0.020)
Marital status (r:
never been in a
relationship)
In a monogamous
relationship
0.75
(0.20)
1.55
(0.52)
1863037.8***
(728997.8)
0.89
(0.22)
1.81*
(0.55)
In a polygamous
relationship
1.91**
(0.54)
3.53***
(1.34)
3613973.4***
(1855414.2)
1.47
(0.44)
4.27***
(1.47)
Divorced,
separated or
widowed
1.17
(0.34)
2.14**
(0.80)
1844043.1***
(805891.7)
1.06
(0.30)
2.02**
(0.67)
270
Psychological
violence
Physical
violence
Sexual
violence
Social violence Economic
violence
Education (r: none)
Primary school 1.12 1.82** 1.23 1.82** 1.50**
(0.26) (0.51) (0.71) (0.43) (0.29)
Middle school 1.07 1.35 1.70 1.89*** 1.38
(0.23) (0.37) (0.82) (0.39) (0.28)
Secondary school 1.45 1.44 2.55* 1.71** 1.21
(0.38) (0.52) (1.38) (0.41) (0.32)
Technical school 1.67 1.21 0.75 1.45 1.24
(0.75) (0.79) (0.82) (0.77) (0.55)
Higher education 1.40 1.17 0.93 1.12 0.98
(0.59) (0.63) (0.66) (0.39) (0.39)
Rural area 1.31 0.96 1.00 0.97 1.39**
(0.22) (0.18) (0.30) (0.13) (0.22)
Patriarchal Gender
Norms Index
1.07
(0.081)
0.98
(0.092)
0.89
(0.11)
0.99
(0.068)
0.93
(0.059)
Tolerance to Wife-
beating Index
1.01
(0.069)
1.13
(0.088)
1.01
(0.15)
1.14**
(0.072)
1.09
(0.067)
Acceptance of
Women’s Sexual
Autonomy Index
1.16**
(0.078)
1.17*
(0.10)
1.13
(0.14)
1.16***
(0.066)
1.11*
(0.071)
Decision-making
Power Index
within the
household
0.97
(0.064)
0.93
(0.079)
1.14
(0.14)
0.91
(0.056)
0.95
(0.056)
Respondent
consumes alcohol
1.24
(0.26)
1.35
(0.31)
1.22
(0.44)
0.97
(0.19)
1.39*
(0.26)
Exposure to
violence in the
community
1.77***
(0.24)
1.49**
(0.26)
3.64***
(1.00)
1.88***
(0.25)
1.57***
(0.22)
Witnessed
psychological
violence at home
as a child
2.21***
(0.46)
271
Psychological
violence
Physical
violence
Sexual
violence
Social violence Economic
violence
Witnessed
physical violence
at home as a child
1.73*
(0.49)
Witnessed sexual
violence at home
as a child
1.18
(0.66)
Witnessed social
violence at home
as a child
1.97***
(0.36)
Witnessed
economic violence
at home as a child
2.93***
(0.60)
Regions (r:
Western)
Central 1.63* 1.32 4.77*** 0.97 0.61**
(0.47) (0.60) (2.27) (0.26) (0.15)
Greater Accra 1.50 1.11 3.08** 1.28 0.51***
(0.41) (0.44) (1.70) (0.32) (0.13)
Volta 0.96 0.96 2.00 0.99 0.22***
(0.25) (0.39) (1.13) (0.26) (0.063)
Eastern 2.95*** 2.03* 1.72 1.66* 0.89
(0.83) (0.76) (0.92) (0.44) (0.20)
Ashanti 1.66* 1.68 1.56 1.20 0.70*
(0.49) (0.65) (0.94) (0.29) (0.15)
Brong Ahafo 1.21 1.19 0.55 0.70 0.70
(0.34) (0.48) (0.37) (0.21) (0.16)
Northern 0.70 0.68 0.65 1.13 0.26***
(0.25) (0.31) (0.53) (0.33) (0.079)
Upper East 0.11** 0.14* 0.97 1.10 0.22***
(0.11) (0.15) (1.09) (0.41) (0.087)
Upper West 0.71 1.07 1.64 1.07 0.47*
(0.41) (0.51) (1.76) (0.35) (0.18)
Observations 2,979 2,879 2,979 2,979 2,979
F-test for regional
effects (p-value)
0.14 0.67 0.31 0.75 0.54
Odds ratio; Standard errors in parentheses are clustered within enumeration areas (EA). * p < 0.1; **
p < 0.05; *** p < 0.01
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
272
Table E2: Determinants of men having experienced domestic violence in the 12 months
prior to the survey: odds ratios estimated using logistic regressions
Psychological
violence
Physical
violence
Sexual
violence
Social
violence
Economic
Violence
Age categories (r:
15–19)
20–24 0.73 0.37** 1.73 0.61 0.86
(0.24) (0.16) (1.69) (0.20) (0.24)
25–29 0.64 0.26** 1.49 0.44** 0.70
(0.25) (0.14) (1.55) (0.18) (0.26)
30–39 0.37** 0.051*** 1.88 0.24*** 0.24***
(0.16) (0.046) (1.87) (0.10) (0.11)
40–49 0.48* 0.059*** 0.52 0.20*** 0.52
(0.20) (0.050) (0.63) (0.088) (0.23)
50–60 0.30** 0.12*** 0.17*** 0.27**
(0.16) (0.092) (0.086) (0.15)
Work situation (r:
self-employed)
Employed 0.91 0.97 2.69* 1.01 2.03***
(0.22) (0.46) (1.59) (0.24) (0.52)
Not working 1.32 0.97 1.68 0.79 2.25**
(0.40) (0.41) (1.22) (0.23) (0.71)
Asset Index 0.99 0.94 1.03* 1.02** 1.00
(0.0089) (0.055) (0.016) (0.0089) (0.019)
Marital status (r:
never been in a
relationship)
In a monogamous
relationship
1.09
(0.29)
0.75
(0.32)
6.02*
(6.18)
2.97***
(0.85)
1.03
(0.30)
In a polygamous
relationship
1.55
(0.66)
0.91
(0.67)
17.4**
(19.9)
5.23***
(2.17)
0.52
(0.33)
Divorced,
separated or
widowed
1.19
(0.37)
0.41
(0.27)
7.91*
(8.51)
1.77*
(0.61)
0.99
(0.34)
273
Psychological
violence
Physical
violence
Sexual
violence
Social
violence
Economic
violence
Education (r: none.
For sexual
violence, r:
Higher)
Primary school 1.87* 1.97 2064392.7*** 1.94 2.87*
(0.68) (1.66) (4702352.7) (0.97) (1.62)
Middle school 1.32 2.18 9448049.2*** 1.73 2.96**
(0.46) (1.63) (26877735.0) (0.74) (1.43)
Secondary 1.91* 4.00* 7752342.1*** 1.54 2.97**
(0.71) (3.12) (16778822.1) (0.71) (1.46)
Technical 1.20 3899351.4*** 2.16 3.70*
(0.73) (9933915.0) (1.35) (2.48)
Higher 1.64 1.81 6917039.8*** 2.23 2.56*
(0.69) (1.91) (3946834.7) (1.09) (1.35)
Rural area 1.28 2.18** 1.18 1.27 1.32
(0.27) (0.75) (0.57) (0.30) (0.31)
Patriarchal Gender
Norms Index
0.92
(0.088)
1.06
(0.22)
0.85
(0.18)
0.84*
(0.083)
1.01
(0.098)
Tolerance to Wife-
beating Index
1.21**
(0.12)
1.47***
(0.19)
1.18
(0.28)
1.04
(0.11)
1.30**
(0.14)
Acceptance of
Women’s Sexual
Autonomy Index
0.91
(0.082)
0.99
(0.16)
1.27
(0.34)
0.96
(0.095)
1.09
(0.11)
Decision-making
Power Index
within household
1.06
(0.13)
1.08
(0.22)
0.85
(0.25)
0.65***
(0.071)
0.88
(0.10)
Respondent
consumes alcohol
1.66***
(0.32)
1.86
(0.72)
2.04*
(0.77)
1.28
(0.26)
1.19
(0.30)
Exposure to
violence in the
community
1.96***
(0.38)
2.26**
(0.77)
2.83***
(1.07)
3.40***
(0.72)
2.62***
(0.51)
274
Psychological
violence
Physical
violence
Sexual
violence
Social
violence
Economic
Violence
Witnessed
psychological
violence at home
as a child
1.80**
(0.51)
Witnessed
physical violence
at home as a child
1.28
(0.58)
Witnessed sexual
violence at home
as a child
2.00
(1.52)
Witnessed social
violence at home
as a child
2.29***
(0.65)
Witnessed
economic violence
at home as a child
2.28***
(0.70)
Regions (r:
Western)
Central 1.44 2.17 1.21 0.79 1.68
(0.53) (1.60) (0.82) (0.28) (0.64)
Greater Accra 1.25 2.39 0.72 1.00 0.80
(0.46) (1.92) (0.49) (0.34) (0.27)
Volta 0.93 0.79 0.60 0.35** 0.20***
(0.32) (0.59) (0.42) (0.15) (0.11)
Eastern 2.43** 2.65 0.44 0.89 1.25
(0.86) (1.96) (0.36) (0.33) (0.47)
Ashanti 1.13 1.51 0.25 0.86 0.89
(0.43) (1.13) (0.21) (0.30) (0.32)
Brong Ahafo 0.35** 1.00 0.27*** 0.86
(0.18) (0.88) (0.12) (0.37)
Northern 0.87 0.24 0.95 0.33** 0.73
(0.35) (0.29) (0.79) (0.15) (0.36)
Upper East 0.35* 1.41 0.16** 0.27*
(0.22) (1.22) (0.12) (0.20)
Upper West 1.48 0.46 0.25
(1.54) (0.26) (0.28)
Observations 1,923 1,904 1,425 1,974 1,974
F-test for regional
effects (p-value)
0.33 0.70 0.78 0.35 0.09
Odds ratio; Standard errors in parentheses are clustered within enumeration areas (EA). * p < 0.1; **
p < 0.05; *** p < 0.01
Source: Ghana Family Life and Health Survey (GFLHS) 2015.
275
Questionnaire annex
276
Q1: Household questionnaire
Administration Form
IDENTIFICATION
ADDRESS _______________________________________________________
LOCALITY NAME ________________________________________________
NAME OF FIRST PRINCIPAL RESPONDENT ________________________
CLUSTER NUMBER_______________________________________________
STRUCTURE NUMBER____________________________________________
HOUSEHOLD NUMBER___________________________________________
REGION_________________________________________________________
DISTRICT________________________________________________________
URBAN =1 | RURAL = 2
_________________________________________________________________
PERSON INTERVIEWED: Man = 1 | Woman = 2
_________________________________________________________________
INTERVIEWER VISITS
1 2 3 FINAL VISIT
DATE
___________
____________
____________
DAY
MONTH
YEAR
INT.NUMBER
RESULT
INTERVIEWER’S
NAME
SUPERVISOR’S
NAME
RESULT
___________
___________
___________
____________
____________
___________
____________
____________
___________
NEXT VISIT:
DATE
TIME
___________
___________
____________
____________
TOTAL
NUMBER
OF VISITS
RESULT CODES:
1 COMPLETED
2 NO HOUSEHOLD MEMBER AT HOME OR NO ELIGIBLE
RESPONDENT AT HOME AT TIME OF VISIT
3 ENTIRE HOUSEHOLD ABSENT FOR EXTENDED
PERIOD OF TIME
4 POSTPONED
5 REFUSED
6 DWELLING VACANT OR ADDRESS NOT A DWELLING
7 DWELLING DESTROYED
8 DWELLING NOT FOUND
9 OTHER ___________________________
TOTAL PERSONS
IN HOUSEHOLD
TOTAL ELIGIBLE
WOMEN
TOTAL ELIGIBLE
MEN
LANGUAGE OF
RESPONDENT
LANGUAGE OF
INTERVIEW
277
LANGUAGE CODES:
ENGLISH = 1 | AKAN = 2 | GA = 3 | EWE = 4 | NZEMA = 5 | DAGBANI = 6 |
OTHER = 7
Informed consent form
Hello, I am [name], and I am working for Ghana Statistical Service. We are conducting a
national survey about family life and women’s/men’s health.
Your household has been randomly chosen to participate in this survey; and we would
like to ask you and another household member some questions. I want to assure you that
all of your answers will be kept strictly secret. I will not keep a record of your name or
address.
I am using a tablet computer to make the interview easier and quicker, but I will not
record anything or take any pictures.
You have the right to stop the interview at any time, or to skip any questions that you don’t
want to answer. There are no right or wrong answers. We would very much appreciate
your household’s participation in this survey.
The interview takes approximately 45 minutes to complete.
Do you have any questions for me?
May we have your permission to ask you questions,
and are you willing to participate?
UNPROMPTED
YES………………………….1
YES, BUT NOT RIGHT
NOW………………………..2
NO…………………………..0
If no, May I ask why you are not willing to participate?
DO NOT ASK ANY FURTHER QUESTIONS TO
THE RESPONDENT AND THANK HIM/HER FOR
HIS/HER TIME.
1. I don’t have the time
2. My spouse/partner/guardian
does not let me
3. I do not want to talk about my
health and life experiences
4. I do not want to talk to you
5. Other
If 2, when would be more suitable for you?
DO NOT ASK ANY FURTHER QUESTIONS TO
THE RESPONDENT AND THANK HIM/HER FOR
HIS/HER TIME.
_________________ Date
_________________ Time
278
TO BE COMPLETED BY INTERVIEWER
I certify that I have read the above consent procedure to the participant.
SIGNED: ___________________________________________________
279
1. Household Roster
READ: First of all, I would like to make a comprehensive list of all members of this household. I would therefore like to ask you to give me the names of all
individuals who normally live and eat their meals together here. Please do not include people who have a household elsewhere and come here only for some hours
during the day or to visit temporarily, e.g. servants and visiting family members or friends. Let us start with the head of the household. MAKE SURE THAT
RESPONDENT IS THE HEAD OF THE HOUSEHOLD!
Special Cases: Domestic Servants if they sleep 5 nights a week or more in the household.
Visitors if they have slept in the household for the past 4 weeks.
Please circle ELIGIBLE individual respondents: ANY WOMAN/MAN 15–60 YEARS OF AGE, LIVING IN THE HOUSEHOLD
Q1: Q2: Q3: Q4: Q5: Q6: Q7: Q8:
Person
ID
Name How old is
[name]?
AGE IN
COMPLETED
YEARS
Circle those who
are eligible for the
individual-level
survey
What is
[NAME]’s
sex?
MALE = 1
FEMALE
= 2
What is
[name]’s
relationship
to the head
of the
household?
SEE CODES
BELOW
What is
[NAME]’s
ethnicity?
98=Don’t
know
What is
[NAME]’s
religion?
Is [NAME]
currently
away from
this
household?
Yes = 1
No = 0
THIS COULD
BE FOR
WORK,
EDUCATION
OR
SPENDING
TIME WITH
FAMILY
Can
[NAME]
read or
write in
any
language?
Yes = 1
No = 0
ASK IF
AGE IS 11
OR
OLDER
1
2
3
4
5
6
7
280
8
9
10
11
12
13
14
15
16
17
What is the ID of the respondent? ______________
CODES: Relationships to head Q4
1 = Head
2 = Spouse
3 = Partner
9 = Brother or sister
10 = Niece/nephew by blood
11 = Niece/nephew by marriage
4 = Son or daughter 12 = Other relative
5 = Son-in-law or daughter-in-law 13 = Adopted/foster/step child
6 = Grandchild 14 = Not related
7 = Parent 98 = Don’t know
8 = Parent-in-law
CODES: Religion Q6
1 = No religion
2 = Catholic
3 =
Anglican/Methodist/Presbyterian
4 = Pentecostal/Charismatic
5 = Other Christian
6 = Islam
7 = Traditional/Spiritualist
8 = Buddhist
9 = Other
CODES: Ethnic group Q5
1 = Akan
2 = Ga/Dangme
3 = Ewe
4 = Guan
5 = Mole-Dagbani
6 = Grussi
7 = Gruma
8 = Mande
9 = Other
CODES: Education level Q10
1 = Pre-school
2 = Primary
3 = Middle/JSS/JHS
4 = Secondary/SSS/SHS
5 = TECH/VOC
6 = Higher
98 = Don’t know
CODES: Work situation Q12
1 = Self-employed
2 = Employee
3 = Casual work
4 = Not working and not looking
5 = Not working and looking for work
6 = Student/pupil
7 = Apprentice
8 = Retired/Too old to work
9 = Disabled/sick
10 = Too young to work
98 = Don’t know
281
CODES: Occupation Q13
1 = Managers
2 = Professionals
3 = Technicians and associate
professionals
4 = Clerical support workers
5 = Service and sales workers
6 = Skilled agricultural, forestry and
fishery workers
7 = Craft and related trade
workers
8 = Plant and machine
operators and assemblers
9 = Elementary occupations
10 = Other occupations
CODES: Marital Status Q14
1 = Never married
2 = Living together but not married
3 = Married
4 = Divorced
5 = Separated
6 = Widowed
282
Q9: Q10: Q11: Q12: Q13: Q14: Q15:
Person
ID
Has [NAME] ever
attended school
or is [NAME]
attending school
now?
1= Never Q11
2 = Now
3 = Past
ASK IF AGE IS 3
OR OLDER
What is the
highest level
of school
[NAME] has
attained?
SEE CODES
ABOVE
How many
years of
schooling has
[NAME]
completed at
this level?
ASK IF AGE
IS 3 OR
OLDER
What describes
[NAME]’s
current main
work situation
best?
SEE CODES
ABOVE
ASK IF AGE IS 5
OR OLDER
FOR
INTERVIEWER:
What describes
[NAME]’s main
current occupation
best?
SEE CODES
ABOVE
ASK IF AGE IS 5
OR OLDER
DO NOT READ
OUT LOUT THE
CODES
What is
[NAME]’s
marital
status?
SEE CODES
ABOVE
ASK IF AGE
IS 12 OR
OLDER
How many years has
[NAME] lived in this
locality?
FULL NUMBER OF
YEARS
HAS ALWAYS LIVED
HERE = 991
LESS THAN ONE YEAR
= 992
998 = Don’t know
1
2
3
4
5
6
7
8
9
10
11
12
13
14
283
15
16
17
18
LIST THE IDs OF ALL THE PERSONS IN THE HOUSEHOLD WHO ARE AT LEAST 13 YEARS OLD
2. Employment
Q1: Q2: Q3: Q4: Q5: Q6: Q7: Q8:
Person
ID
Has [NAME]
done any
work over
the last 12
months?
1 = Yes
0 = No
Q13
Over the last
12 months,
how many
months was
[NAME]
involved in
any economic
activities?
EXCLUDING
UNPAID
HOUSEWORK
OR CHORES
Did [NAME]
work during the
last 7 days?
1 = Yes
0 = No
EXCLUDING
UNPAID
HOUSEWORK
OR CHORES
IF NO >>>
NEXT PERSON
IF LAST
PERSON AND
NO>>Q13
Please describe
what work
[NAME] has done
over the last 7
days.
For interviewer:
What work
activities was
[NAME] involved
in during the last
7 days?
SEE
OCCUPATION
CODES
DO NOT READ
OUT LOUT THE
CODES
On
average,
how many
hours per
day was
[NAME]
engaged
in all work
activities
during the
last 7
days?
Was [name]
paid in
cash, in-
kind (self-
consumptio
n) or as a
contributin
g family
member for
any of these
activities?
1 = Yes
0 = No
If No >>
Q11
How was
[NAME] paid
for all the
work
activities
during the
last 7 days?
(multiple
response)
1 = In kind
2 = Cash
3 =
Contributing
family
member
IF 2 >> Q10
If 3 >> Q11
Main
activity
Second
ary
activity
Main
activity
Second
ary
activity
284
1
2
3
4
5
6
7
8
9
10
11
12
Q9: Q10: Q11: Q12: Q13: Q14:
Person
ID
What did
[NAME] receive
in-kind for all
work activities
during the last
month?
1 = Food
2 =
Accommodation
or lodging
3 = Education
(schooling or
vocational)
4 = Clothes
5 = Medicine
6 = Other
How much did [NAME]
earn for all work
activities?
Per:
1 = Day
2 = Week
3 = Month
IF PAID IN-KIND, WHAT
IS THE VALUE?
Who was [NAME]
working for during the
last 7 days? Prompted
1 = Employed by family
member
2 = Private employer
3 = Public sector
4 = Self-employed
within the family
5 = Self-employed
outside the family
6 = Friend
7 = Other
2 ANSWERS
ALLOWED;
Did [NAME] have
a work contract
for the main
activity?
1 = Formal
contract
2 = Verbal contract
0 = No contract
Was [NAME]
involved in house
work/chores (e.g.
cleaning, laundry,
taking care of
children, cooking)
during the last 7
days?
1 = Yes
0 = No
IF NO>>>NEXT
INDIVIDUAL
On average, how
many hours per
day did [NAME]
work on house
work/chores
during the last 7
days?
THEN MOVE TO
NEXT
INDIVIDUAL
285
Main
occupation
Secondary
occupation
Main
occupation
Secondary
occupation
Cash Kind Cash Kind
1
2
3
4
5
6
Q16: Over the last 12 months, has anyone in your household lost their job? 1 = Yes
0 = No >>>then next module
Q17: If yes, then who?
Use household IDs
3. Food consumption and household shocks
Thank you for this information. Now, I would like to know a little bit more about what your household consumes for food.
Name of the item a) Did your household consume
[ITEM] in the last 7 days?
Yes…1
No….0 >>>> Next item
b) On how many days over
the last 7 days did your
household consume this
[ITEM]?
Q1: Cereals
Q2: Beans/pulses/lentils
Q3: Edible oil
Q4: Leafy vegetables
Q5: Any other vegetables
Q6: Tubers (yam, cocoyam, cassava etc.)
Q7: Plantain
Q8: Dairy product, milk
Q9: Eggs
Q10: Meat
286
Name of the item a) Did your household consume
[ITEM] in the last 7 days?
Yes…1
No….0 >>>> Next item
b) On how many days over
the last 7 days did your
household consume this
[ITEM]?
Q11: Fruits
Q12: Fish
Q13: Spices
Q14: Other food (e.g. sugar, honey, tea leaves, coffee)
Q15: During the last lean season, how often have you or a member of your household
gone to bed while still hungry?
1 = more than once a week
2 = once a week
3 = a few times per month
4 = once a month
5 = rarely
6 = never
Q16: During the last harvest season, how often have you or a member of your
household gone to bed while still hungry?
1 = more than once a week
2 = once a week
3 = a few times per month
4 = once a month
5 = rarely
6 = never
Now I would like to ask you some questions related to income shocks that your household might have experienced in the last 12 months.
Over the last 12 months, have you or anyone in your household been affected by any
of the following things?
1 = Yes 0 = No
Q17: Flooding/too much rain
Q18: Drought
Q19: Landslides
Q20: Pest or diseases (crops or livestock)
Q21: Fire (burning down property)
Q22: Insufficient water supply for farming or gardening
287
Q23: How would you or other members of the
household get 1000 Cedis if you really needed to.
For example, because of a medical emergency,
MULTIPLE ANSWERS ALLOWED
1 = Taking a formal loan (from e.g. bank)
2 = Taking an informal loan (from e.g. moneylenders, farmers)
3 = Asking relatives/friends in the same village/town
4 = Asking relatives/friends in different location
5 = Use savings
6 = Use microfinance
7 = Sell durable goods/equipment
8 = Sell land/house
9 = Sell livestock
10 = Take extra work
11 = Other
288
4. Household Assets and Decision Making
Now I would like to know about goods that belong to you or any other household member. Please also consider goods that are part of your
business.
How many of the following things does your household possess
and who (mainly) owns and uses the respective item(s)?
A. OWN
Yes = 1, No = 0
B. Owner(s): Use hhd IDs
MULTIPLE IDs ALLOWED
Q1: Bicycle
Q2: Motorcycle or motor scooter
Q3: Tricycle
Q4: Animal-drawn cart
Q5: Car
Q6: Truck
Q7: Boat with a motor
Q8: Boat without a motor
Q9: Tractor
Q10: A) How much agricultural land does your household possess?
If 0, skip to Q13
B) Unit:
1= Hectares
2 = Acres
3 = Poles
4 = Plots
C) Who owns the agricultural land? Household IDs
D) Who usually make decisions regarding the use of agricultural land
THIS INCLUDES DECISIONS ON WHAT TO GROW AND SALE AND RENT
Household IDs
Q11: Is part of this land cultivated by a tenant? Yes=1, No=0 >>> Q13
Q12: For the plots which you did not crop yourself, what net rent did
you receive from the tenant during the last year? IF RENT IS
PAID IN KIND, ASK THE RESPONDENT TO ESTIMATE THE
MONETARY VALUE
Total amount in Cedis
____________________
98 = Don’t know
Q13: How much agricultural land does your
household cultivate?
1= Hectares
2 = Acres
3 = Poles
289
INCLUDE CULTIVATED LAND THAT IS
RENTED IN AND EXCLUDE LAND THAT IS
RENTED OUT. If 0 >> NEXT MODULE
4 = Plots
Q14: How much of this cultivated land is irrigated?
If 0 >>Q16
1= Hectares
2 = Acres
3 = Poles
4 = Plots
Q15: What is the principal source of
water for this agricultural land?
1 = Dam
2 = Hole with a pump
3 = Well
4 = Natural spring
5 = River/lagoon/lake
6 = Storage tank
7 = Rainfall
8 = Other
Q16: IF HOUSEHOLD OWNS AND CULTIVATE LAND, What were
your net earnings of the last 12 months from the land you own
and cultivate? IF HARVEST IS NOT SOLD TO THE MARKET,
ASK THE RESPONDENT TO ESTIMATE THE MONETARY
VALUE
Total amount in Cedis
______________________
98 = Don’t know
Q17: Does your household cultivate land owned by someone outside
your household?
1=Yes
0=No >>> NEXT MODULE
Q18: What was the net earnings of the last 12 months for this rented
land? IF HARVEST IS NOT SOLD TO THE MARKET, ASK THE
RESPONDENT TO ESTIMATE THE MONETARY VALUE
Total amount in Cedis
________________________
98 = Don’t know
For agricultural purposes: IF HOUSEHOLD DOES NOT OWN OR CULTIVATE LAND, SKIP TO Q32
How many of the following things does your household possess and who
(mainly) owns the respective item(s)?
A. How many? B. Owner(s): Use hhd IDs
MULTIPLE IDs ALLOWED
Q19: Plough
Q20: Cart
Q21: Thresher
Q22: Trolley
Q23: Water pump
Q24: Irrigation pump
Q25: Grain storage bin
Q26: Sprayers
290
Q27: Cattle
Q28: Milk cow and bulls
Q29: Horses, donkeys, mules
Q30: Goats
Q31: Sheep
Q32: Pigs
Q33: Rabbits
Q34: Grasscutter
Q35: Poultry
Q36: Who in your household makes decisions regarding the use of the following animals:
a) Cattle
b) Milk cows and bulls
c) Horses, donkeys and mules
d) Goats
e) Sheep
f) Pigs
g) Rabbits
h) Grasscutter
i) Poultry
Q37: Does anyone in your household have a bank account? Who? Yes = 1, No = 0
Q38: Does anyone in your household use a microfinance scheme? Who? Yes = 1, No = 0
Q39: Do you or any member of your household receive remittances from people
living outside your household?
Yes = 1
No = 0 >>>Q41
Q40: What is the value of these remittances? ____________________
1 = Day 3 = Month
2 = Week 4 = Year
Q41: Are you or any member of your household a participant in or beneficiary of a
cash transfer programme by the government? (LIST) THIS DOES NOT
INCLUDE CASH TRANSFERS FROM FRIENDS OR FAMILY
Yes = 1
No = 0 >>Q43
Q42: What is the value of the transfer? _______________________
291
THIS DOES NOT INCLUDE CASH TRANSFERS FROM FRIENDS OR FAMILY 1 = Day 3 = Month
2 = Week 4 = Year
I would like to ask you about spending decisions in your household.
How are spending decisions made in your household? Who usually… Decision-maker (Use Household IDs)
0 = None of us
199 = All of us
98 = Don’t know
99 = Not applicable
Q43: Decides how the earnings of the head of the household are used?
Q44: Decides how the money from the remittances is used?
Q45: Decides how the money from the cash transfer programme is used?
Who decides how to spend money on the following items?
Q46: Food
Q47: Education for the children
Q48: Healthcare for the children
Q49: Healthcare for yourself
Q50: Clothing
Q51: Purchase for daily household needs
Q52: Major household purchases
Q53: Do you own or rent the dwelling that you
live in?
1= Owner occupied
2 = Renting >>>>>Q55
3 = Rent free >>>>>Q55
4 = Perching >>>>>Q55
5 = Squatting >>>>>Q55
Q54: Who owns it? Roster IDs
Q55: H
0
1
I would like to ask you a few questions
about your housing condition.
What is the main source of drinking water
for members of your household?
1 = Piped into dwelling/indoor
2 = Piped to yard/plot
3 = Public tap/standpipe
4 = Tube well or borehole
5 = Protected well
6 = Unprotected well
7 = Protected spring
9 = Rainwater
10 = Tanker truck
11 = Cart with small tank
12 = Surface water (river/dam/lake/pond
/stream/canal/irrigation channel)
13 = Bottled water
14 = Sachet water
292
8 = Unprotected spring 15 = Other
Q56: Which of the following does your
household use the most:
1 = A private toilet
2 = A shared toilet with other families
3 = A public toilet
4 = None of these
Q57: H
0
2
What main kind of toilet
facility do members of your
household usually use?
1= Flush to piped sewer system
2 = Flush to septic tank
3 = Flush to pit latrine
4 = Flush to somewhere else
5 = Flush, don’t know where
6 = Ventilated improved pit latrine
7 = Pit latrine with slab
8 = Pit latrine without slab/open pit
9 = Bucket/pan
10 = Composting toilet
11 = No facility/bush/field/sea
Q58: H
0
3
Which material is mainly used
for the roof of your house?
1 = No roof
2 = Thatch/palm leaf/sod
3 = Palm/bamboo
4 = Wood planks
5 = Cardboard
6 = Metal
7 = Wood
8 = Calamine/cement fibre
9 = Ceramic tiles/brick tiles
10 = Burnt bricks
11 = Cement
12 = Roofing shingles
13 = Asbestos/slate
14 = Other
Q59: Which material is mainly used
for the exterior walls of your
house?
1 = No walls
2 = Cane/palm/trunks
3 = Dirt/mud
4 = Bamboo with mud
5 = Stone with mud
6 = Uncovered adobe
7 = Plywood
8 = Cardboard
9 = Reused wood
10 = Cement or cement blocks
11 = Stone with lime/cement
12 = Bricks
13 = Covered adobe
14 = Wood planks/shingles
Q60: Which material is mainly used
for the floor of your house?
1 = Earth/sand
2 = Dung
3 = Wood planks
4 = Palm/bamboo
5 = Parquet or polished wood
6 = Vinyl or asphalt strips
7 = Ceramic tiles/terrazzo
8 = Cement
9 = Woollen carpet/synthetic carpet
10 = Linoleum/rubber carpet
Q61: How many rooms do you have in the dwelling that you live in?
Q62: How many rooms do you use for sleeping in the dwelling that you live in?
293
5. Household Assets and Problems in the Community
Now I would like to ask you about your source(s) of electricity.
On a typical day, how many hours
per day do you use this source of
energy for lighting?
Q24: Grid electricity (household is a utility
consumer)
Q25: Generator (from neighbour or relative)
Q26: Generator (personal)
Q27: Mini-hydro
Q28: Solar home system (e.g. solar panel)
Q29: Storage (car) battery
Q30: Candles
Q31: Kerosene (wick and hurricane lamps)
Q32: LPG
Finally, I would like to ask you some questions about your community.
Q33: On a typical day, in your community, how many hours of electricity
is available?
Do you know if any of the following are problems in your
community or neighbourhood?
Yes = 1 No = 0
Q34: Chieftaincy disputes
Q35: Sales of drugs (illicit or hard drugs)
Q36: Drug abuse (illicit or hard drugs)
Q37: Alcohol abuse
Q38: Domestic Violence
Q39: U Poor/unreliable water provision
Q40: Poor/unreliable electricity provision
Q41: Poor/unreliable waste removal
Q42: Poor roads
Q43: Poor health service provision
Q44: Is there a shortage of water in this community
during the dry season?
1 = Yes
0 = No >>> Q52
2 = There is always dry season >>>Q49
Does your household have: 1 = Yes
0 = No
1 = Yes
0 = No
Q1: Radio Q2: TV
Q3: Mobile phone Q4: Bed
Q5: Other furniture Q6: Cooking utensils
Q7: Washing machine Q8: Fan
Q9: Air conditioner Q10: Refrigerator
Q11: Iron Q12: Sewing machine
Q13: Kerosene stove Q14: Gas/LPG/biogas stove
Q15: Electric stove Q16: Open stove/open fire/coal pot
Q17: Dish/cable TV Q18: Wall clock
Q19: Landline telephone Q20: Digital photo-camera
Q21: Tablet Q22: Computer
Q23: Mosquito nets
294
1 = More
than
once a
week
2 =
Once
a
week
3 = A
few
times a
month
4 =
Once a
month
5 =
Rarely
(once in
a
while)
6 =
Never
Q45: How often is the shortage of
water in this community
during the dry season?
Q46: How often do disputes over
water occur between
members of this community
during the dry season?
Q47: How often do disputes over
water occur between
members of this community
and a different community
during the dry season?
Q48: How often is the shortage of
water in this community?
Q49: How often do disputes over
water occur between
members of this
community?
Q50: How often do disputes over
water occur between
members of this community
and a different community?
Q51: Is there a shortage of agricultural land in this
community?
1 = Yes, 0 = No >>> Q56,
99 = N.A. >>> Q56
Q52: How often do disputes over
agricultural land occur
between members of this
community?
Q53: How often do arguments
over grazing land occur in
this community? If not
applicable, because there is
no grazing land, enter 99
Q54: How often do arguments
over grazing land occur
between members of
different communities? If not
applicable, because there is
no grazing land, enter 99
Q55: Are mobile phone networks available
in all areas of this community?
1 = All
2 = Most
3 = Some
4 = Few
5 = None
295
Q56: Which method of transport would you
primarily use to go to the district capital?
CIRCLE THE APPROPRIATE AND
MAIN MEAN OF TRANSPORT
1 = Foot
2 = Car
3 = Bicycle
4 = Motorcycle
5 = Tricycle
6 = Boat/canoe
7 = Public transport (e.g.
bus or tro-tro)
8 = Truck
9 = Animal drawn cart
10 = Tractor
98 = Don’t know
Q57: How long does it take to go to the district capital using this
method of transport in minutes?
1 = Minutes
2 = Hours
Thank you very much for your time. I would like to interview another member of your household
to go into more detail about issues related to health and society.
Could I take your phone number for verification purposes? __________________________
Situation of respondent:
1 Alone
2 Children were present some of the time (6–14 years old)
3 Children were present all of the time (6–14 years old)
4 Adults were present some of the time (above 14 years old)
5 Adults were present all of the time (above 14 years old)
66 Other
296
Selection of Respondent for the Individual Interview
Look at the last digit of the household questionnaire serial number on the cover page. This is the
row number you should go to. Check the total number of eligible females or males in the household
schedule. How many are eligible? ___________
This is the columnyou should go to. The cell where the row and the column meet is the number
of the selected woman or man for the domestic violence module in the household schedule.
For example, the household was selected to interview a woman and there are three eligible women
age 15–60 (line IDs 02, 04 and 05). If the household questionnaire serial number is ‘216’, the last
digit is ‘6’; therefore, go to row ‘6’. There are three eligible women age 15–60 in the household,
therefore go to column ‘3’. Follow the row and column and find the number where the row and
column meet (‘2’) and circle the box. Now go to the household schedule and find the second woman
who is eligible for the woman's interview (line number ‘04’ in our example). Write her line ID and
name (if given) below.
TABLE FOR SELECTION OF RESPONDENT
LAST DIGIT OF
THE HOUSEHOLD
QUESTIONNAIRE
SERIAL NUMBER
TOTAL NUMBER OF ELIGIBLE WOMEN 15–60 /
MEN 15–60 IN THE HOUSEHOLD
1 2 3 4 5 6 7 8
0 1 2 2 4 3 6 5 4
1 1 1 3 1 4 1 6 5
2 1 2 1 2 5 2 7 6
3 1 1 2 3 1 3 1 7
4 1 2 3 4 2 4 2 8
5 1 1 1 1 3 5 3 1
6 1 2 2 2 4 6 4 2
7 1 1 3 3 5 1 5 3
8 1 2 1 4 1 2 6 4
9 1 1 2 1 2 3 7 5
Line ID of selected woman/man: _________
Name of selected woman/man: ___________________________________________________
ASK IF YOU CAN TALK WITH THE SELECTED WOMAN/MAN for the second part of the
questionnaire. IF SHE/HE IS NOT AT HOME, AGREE ON A DATE FOR A RETURN VISIT.
297
Q2: Individual questionnaire Administration Form
IDENTIFICATION
ADDRESS _________________________________________________________
LOCALITY NAME _________________________________________________
NAME OF FIRST PRINCIPAL RESPONDENT _________________________
CLUSTER NUMBER _______________________________________________
STRUCTURE NUMBER ____________________________________________
HOUSEHOLD NUMBER ___________________________________________
REGION _________________________________________________________
DISTRICT ________________________________________________________
URBAN =1 | RURAL = 2
_________________________________________________________________
PERSON INTERVIEWED: Man = 1 | Woman = 2
_________________________________________________________________
INTERVIEWER VISITS
1 2 3 FINAL VISIT
DATE
___________
____________
____________
DAY
MONTH
YEAR
INT. NUMBER
RESULT
INTERVIEWER’S
NAME
SUPERVISOR’S
NAME
RESULT
___________
___________
___________
____________
____________
___________
____________
____________
___________
NEXT VISIT:
DATE
TIME
___________
___________
____________
____________
TOTAL
NUMBER
OF VISITS
RESULT CODES:
1 COMPLETED
2 THE FIRST RESPONDENT DID NOT AGREE TO THE INTERVIEW OF THE SECOND
RESPONDENT
3 REFUSED BY THE SECOND RESPONDENT
4 SECOND RESPONDENT NOT AVAILABLE
5 COULD NOT PROCEED BECAUSE OF LACK OF PRIVACY
6 OTHER ___________________________
LANGUAGE OF
RESPONDENT
LANGUAGE OF
INTERVIEW
298
LANGUAGE CODES:
ENGLISH = 1 | AKAN = 2 | GA = 3 | EWE = 4 | NZEMA = 5 | DAGBANI = 6 |
OTHER = 7
Was the respondent alone during the following domestic violence
modules? 1 = Yes 0 = No
Social violence or controlling behaviour
Sexual violence
Physical violence
Psychological violence
Economic violence
299
Hello, I am [name], and I am working for Ghana Statistical Service. We are conducting a
national survey about women’s/men’s health and life experiences.
Your household has been randomly chosen to participate in this survey; and we would
like to ask you some questions.
I want to assure you that all of your answers will be kept strictly secret. I will not keep a
record of your name or address. You have the right to stop the interview at any time, or to
skip any questions that you don’t want to answer. There are no right or wrong answers.
Some of the topics may be difficult to discuss, but many women/men have found it useful
to have the opportunity to talk.
We would very much appreciate your participation in this survey.
Do you have any questions?
(The interview takes approximately 1 hour and 30 minutes to complete.)
May I have your permission to ask you questions, and
are you willing to participate?
YES………………………….1
NOT NOW………………….2
NO…………………………..0
If no, May I ask why you are not willing to participate?
DO NOT ASK ANY FURTHER QUESTIONS TO
THE RESPONDENT AND THANK HIM/HER FOR
HIS/HER TIME.
1. I don’t have any time
2. My spouse/partner/guardian
does not let me
3. I do not want to talk about my
health and life experiences
4. I do not want to talk to you
5. Other
If it is not a good time to talk, when would be more
suitable for you?
DO NOT ASK ANY FURTHER QUESTIONS TO
THE RESPONDENT AND THANK HIM/HER FOR
HIS/HER TIME.
_________________ Date
_________________ Time
Q1: For interviewers: What is sex of
respondent?
1 = Male, 2 = Female
TO BE COMPLETED BY INTERVIEWER
I certify that I have read the above consent procedure to the participant.
SIGNED: _______________________________________________________
300
1. Safety and Social Capital
First, I would be interested to know a bit more about how you feel living in your community.
Q2: How safe do you feel in the area you
live in?
1 = Very safe
2 = Somewhat safe
3 = Somewhat unsafe
4 = Very unsafe
Q3: Why don’t you feel
safe in the area you
live in?
1. Fear of being robbed
2. Fear of
burglary/vandalism while
being away
3. Fear of sexual
harassment/rape
4. Fear of verbal harassment
5. Fear of physical
harassment/attacked
6. Fear of being insulted/ridiculed
7. Fear of something else
Q4: How safe do you
think women and
girls in this area feel
when going out
alone?
During the day During the night
1 = Very
safe
2 =
Somewhat
safe
3 = Somewhat
unsafe
4 = Very
unsafe
1 = Very safe
2 =
Somewhat
safe
3 = Somewhat
unsafe
4 = Very unsafe
Q5: How safe do you
think men and boys in
this area feel when
going out alone?
During the day During the night
1 = Very
safe
2 =
Somewhat
safe
3 = Somewhat
unsafe
4 = Very
unsafe
1 = Very safe
2 =
Somewhat
safe
3 = Somewhat
unsafe
4 = Very unsafe
Q6: Q7: Q8:
Have you
ever
witnessed
an act
related to
[problem]
over the last
12 months?
1 = Yes,
0 = No
Have you ever
personally
been affected
by or
experienced
acts related to
[problem] over
the last 12
months?
1 = Yes
0 = No >>>
Next item
How were you affected?
1 = Damage to
personal property
2 = Loss of
personal property
3 = Physical harm
4 = Psychological
harm, stress or
worry
5 = Loss of
income
6 = Damage
to communal
property
7 = Fear of
leaving the
house
8 = Other
A) Theft, robbery, burglary
B) Vandalism
C) Arson
D) Kidnapping/abduction
E) Extortion
F) Local bullies
G) Political/communal
violence
H) Highway attacks
I ) Sexual harassment
G) Rape
K) Defilement
301
L) Seduction
Q9: If a group of people in your neighbourhood decided to organise
a neighbourhood project, such as a sanitation or clean-up project,
would you like to contribute (time or money) to this project?
1 = Yes, happily
2 = Yes, I will
have to
0 = No
Q10: How many marriage ceremonies (weddings, engagements) have
you been personally invited to in the last 12 months?
________________
Q11: How many funerals have you attended in the last 12 months? ________________
Q12: Are you a member of any of the following groups, organisations and associations?
Prompt
1 = Area residents’ association
2 = Cooperative
3 = Professional association
4 = Trade union
5 = Credit/finance group
6 = Neighbourhood council
(including landlords’ association)
7 = School committee or parent’s
group (PTA)
8 = Sports group or club,
youth group
9 = NGO or civic group
10 = Wives’ association or
club
11 = Fan/social club
12 = Cultural association (e.g.
theatre group, reading
group)
13 = Political group or
association
14 = Women’s group
15 = Religious group
16 = Family committee
17 = Ethnic or tribal
association
18 = Student
association (Old Boy’s
club)
19 = Other
Q13: If a member of your household were to get into a disagreement with someone from
this community, who would they approach to resolve it?
1 = No one, they will solve it
amongst themselves
2 = Their friends
3 = Family members
4 = Police
5 = Other government
officials
6 = Community
leaders
7 = Chief/Queen
mother
8 = The district
assembly
9 = Religious leaders
10 = Other members of the
community
11 = Other
12 = Don’t know
302
2. Safety in the Home and Controlling Behaviour
Q1: How safe do you feel in your
home?
1 = Very safe
2 = Somewhat safe
3 = Somewhat unsafe
4 = Very unsafe
The next questions are about things
that happen to many men and
women. Please can you tell me if
anyone has ever (even when you
were a child) done or tried to do the
following things to you:
Clarify: Anyone includes your partner,
members of your household, family,
work colleagues etc.
a) Kept you
from seeing
your friends
or family of
birth
b) Stopped you
from leaving
your house?
c) Insisted on
knowing where
you are at all
times including
by controlling,
checking your
phones, texts or
emails
d) Stalked you,
including by
calling,
messaging,
watching or
following you to
a point where
you felt
uncomfortable
e) Spread false
information
about you or
distributed
photos or
videos of you
without your
permission
f) Forced you to
have an
abortion
Q2: 1 = Yes
2 = Someone tried >>> NEXT
ITEM
0 = No >>> NEXT ITEM
Q3: Has this happened in the last
12 months?
1 = Yes, 0 = No >>> NEXT
ITEM
Q4: In the last 12 months, was the
person who did this to you
from your current household?
1 = Yes all of them, 2 = Some
of them, 0 = No >>Q6
Q5: In the last 12 months, who did
this to you from your current
household? Use IDs from the
303
household roster
Then skip to Q8
Q6: In the last 12 months, who did
this to you? Use codes below
Q7: What was the sex of this
person?
1 = Male, 2 = Female
Q8: Over the last 12 months, how
often has this happened? Use
codes below
Q9: When you were a child, did you ever observe this kind of behaviour between
your parents or other members of your household?
1 = Yes, parents
2 = Yes, other members
0 = No
Q10: Think of 10 of your closest neighbours, how many of them do you know, or
think, experienced this kind of behaviour over the last 12 months?
98 = Don’t know
_______________ out of 10
CODES FOR Q6 and 16: WHO DID THIS TO YOU? AND WHO DID YOU DO THIS TO? CODES FOR Q8 and 13: HOW OFTEN HAS
THIS HAPPENED?
304
1 = Partner
2 = Former partner
3 = In-laws/partner’s family
4 = Parents
5 = Sons, daughters
6 = Brothers, sisters
7 = Uncles, aunts
8 = Grandparents
9 = Cousins, nieces, nephews, other
relatives
10 = Priest or religious leader
11 = Other relatives who do not live in this household
but in the community
12 = Other relatives who live outside this community
13 = Friend
14 = Teacher
15 = Employer
16 = Classmates
17 = Work colleagues
18 = Police
19 = Other people in the community
20 = Other people outside the community
1 = Once only
2 = Once in a while
3 = Less than once per
month
4 = Once per month
5 = 2–3 times per month
6 = Once a week
7 = 2–3 times a week
8 = Almost every day
9 = Every day
Please can you tell me if
you have ever done or tried
to do the following things
to someone else:
Clarify: Anyone includes your
partner, members of your
family, work colleagues etc.
a) Kept
someone from
seeing his/her
friends or
family of birth
b) Stopped
someone from
leaving their
home
c) Insisted on
knowing where
someone is at all
times including by
controlling or
checking their
phones, texts or
emails
d) Stalked
someone
including by
calling, texting,
watching or
following to a
point where you
made them feel
uncomfortable
e) Spread false
information about
someone and/or
distributed
photos or videos
of someone
without their
permission
f) Forced someone
to have an abortion
Q11: 1 = Yes
2 = I tried >>> NEXT
ITEM
0 = No >>> NEXT
ITEM
305
Q12: Has this happened in
the last 12 months?
1 = Yes, 0 = No >>>
NEXT ITEM
Q13: Over the last 12
months, how often has
this happened?
Use codes above
Q14: Over the last 12
months, was the
person you did this to
from your household?
1 =Yes, 0 = No
>>>>Q16
Q15: Over the last 12
months, who did you
do this to from your
household? Use
Household IDs
Q16: Over the last 12
months who did you
do this to?
Use codes above
Q17: What was the sex of
this person?
1 = Male, 2 = Female
306
3. Attitudes
I would like to know more about how people sometimes relate to each other and your thoughts
about it. We have prepared 4 short stories and would like to know your reaction. Please keep in
mind that there are no right or wrong answers.
S1: After a marital dispute the wife refused to talk to her husband, and would refuse to make his
dinner or any other household chores. The husband hit his wife, but without leaving a mark.
Q1: How acceptable do you think the
behaviour of the husband is?
1 = Completely
unacceptable
2 = Somewhat
unacceptable
3 = Somewhat
acceptable
4 = Completely
acceptable
98 = Don’t know/Can’t
say
Q2: Do you think it is the woman’s fault? 1 = Yes, mostly
2 = Partially
3 = Not at all
S1’: After a marital dispute the wife refused to talk to her husband, and would refuse to make
his dinner or any other household chores. The husband hit his wife, but without leaving a mark.
The local priest/Imam said that it was the woman’s fault because she didn’t fulfil her duties.
Q1: How acceptable do you think the
behaviour of the husband is?
1 = Completely
unacceptable
2 = Somewhat
unacceptable
3 = Somewhat
acceptable
4 = Completely
acceptable
98 = Don’t know/Can’t
say
Q2: Do you think it is the woman’s fault? 1 = Yes, mostly
2 = Partially
3 = Not at all
Q3: How likely do you think it is that
your local priest/Imam would react
like this?
1 = Very likely
2 = Somewhat likely
3 = Somewhat
unlikely
4 = Very unlikely
98 = Don’t know/Can’t
say
S2: A newly married man in your community, who is usually a very calm and generous man,
has been facing some financial problems. One day his boss informs him that he has lost his job.
He comes home after this, has an argument with his wife, and beats her up to the point that she
passes out/becomes unconscious. This is the first time an incident like this happens. The wife
later reports the incident to the police, and the man is arrested.
Q4: Do you agree that the woman did the
right thing by going to the police?
1 = Fully agree
2 = Somewhat agree
3 = Somewhat
disagree
4 = Fully disagree
98 = Don’t know/Can’t
say
Q5: If 3–4, why do
you think the
woman should
not have gone to
the police?
1 = Because it was the first time it
happened
2 = Because such matters should
remain in the family
3 = Because this is a normal
behaviour
4 = Because it will break the
family
5 = Because it will create
financial problems
6 = Other
307
Q6: Do you agree that the police did the
right thing by arresting the man?
1 = Fully agree
2 = Somewhat agree
3 = Somewhat disagree
4 = Fully disagree
98 = Don’t know/Can’t
say
Q7: If 3–4, why do you
think the police
should not have
arrested the man?
1 = Because it was the first time it
happened
2 = Because such matters should
remain in the family
3 = Because this is a normal
behaviour
4 = Because it will break the
family
5 = Because it will create
financial problems
6 = Because the punishment is
too harsh
7 = Other
Q8: Q
8
:
What do you think
the most adequate
punishment for the
husband would be?
1 = Financial compensation to the
wife or her family
2 = Discussions between the wife
and husband’ s families
3 = Sentencing to jail
4 = Official warning by the
police/justice
5 = Police should be physically
rough with him
6 = Apology
7 = Fine
8 = Husband should be bonded
to a note of good behaviour
9 = It is enough to just arrest
him
10 = No punishment needed
S2’: A woman repeatedly makes demeaning comments to her husband about him not making
enough money, and not being able to provide for a nice home and gifts for her. She constantly
compares him to his brother, who is a successful businessman. One evening, after the woman
made such comments and the couple get into an argument, the man beats her up to the point
that she passes out/becomes unconscious. This is the first time an incident like this has
happened. The wife later reports the incident to the police, and the man is arrested.
Q1: How common do you think it is that a
woman treats her husband like this?
1 = Very common
2 = Somewhat
common
3 = Somewhat
uncommon
4 = Very uncommon
Q2: Do you agree that the woman did the
right thing by going to the police?
1=Fully agree
2= Somewhat agree
3=Somewhat disagree
4=Fully disagree
98 = Don’t know/Can’t
say
Q3: If 3-4, why do you
think the woman
should not have
gone to the police?
1 = Because it was the first time it
happened
2 = Because such matters should
remain in the family
3 = Because this is a normal
behaviour
4 = Because it will break the family
5 = Because it will create
financial problems
6 = Because the woman is the
one to blame
7 = Other
Q4: Do you agree that the police did the
right thing by arresting the man?
1 = Fully agree
2 = Somewhat agree
3 = Somewhat
disagree
4 = Fully disagree
98 = Don’t know/Can’t
say
308
Q5: If 3–4, why do you
think the police should
not have arrested the
man?
1 = Because it was the first
time it happened
2 = Because such matters
should remain in the family
3 = Because this is a normal
behaviour
4 = Because it will break the
family
5 = Because it will create
financial problems
6 = Because the punishment is
too harsh
7 = Because the man is not to
blame
8 = Other
Q6: What do you think the
most adequate
punishment for the
husband would be?
1 = Financial compensation to
the wife or her family
2 = Discussions between the
wife and husband’ s families
3 = Sentencing to jail
4 = Official warning by the
police/justice
5 = Police should be
physically rough with him
6 = Apology
7 = Fine
8 = Husband should be bonded
to a note of good behaviour
9 = It is enough to just arrest him
10 = No punishment needed
S3: Suppose you hear a dispute between your closest neighbour and her young son. You then
witness the son leave the house crying and with several injuries on his face. You know that there
are two other young children in the house, and you can still hear your neighbour screaming.
Q7: What do you
think you should
do?
MULTIPLE
ANSWERS
ALLOWED
1 = Nothing, I don’t think it is a
problem
2 = Nothing, it is not my
business
3 = Approach the police and ask
them to deal with it
4 = Approach religious
authorities and ask them to deal
with it
5 = Approach community
leaders and ask them to deal
with it
6 = Go directly to your neighbour’s
house and try to stop the dispute
and calm down your neighbour
7 = Approach a relative of the
neighbour and ask them to deal
with it
8 = Approach other members of the
community for help.
9 = I would go and check what
happened
S4: Suppose a man in your community makes lewd comments to his niece and sometimes
touches her inappropriately. This man is a community leader, and talks about the girl
inappropriately in front of other people, making comments about her revealing clothes and
body. The girl’s mother asks her to wear more modest clothes whenever the uncle visits but
does nothing further.
Q8: What do you
think the girl
should do?
MULTIPLE
ANSWERS
ALLOWED
1 = She should dress more
modestly
2 = She should stand up for
herself/ask the uncle to stop
3 = She should threaten the uncle
that she will tell the police or other
authorities
4 = She should tell her father
5 = She should tell other family
members
6 = She should tell religious leaders
7 = She should tell the police or
other authorities
8 = The girl should take this as a
compliment
9 = This is what men do; she should
avoid being with him
309
Q9: What do you
think the
mother should
do?
1 = The mother is right to ask her
daughter to dress modestly
2 = She should ask the uncle to stop
3 = She should threaten the uncle
that she will tell the police or other
authorities
4 = She should tell other family
members
5 = She should tell religious leaders
6 = She should tell the police or
other authorities
7 = She should advise her daughter
to go to the police/authorities
8 = She should tell the girl to take
this as a compliment
9 = This is what men do; she should
advise/help her daughter to avoid
being with him
S5: Suppose that someone that you know, who has been dating his girlfriend for two years now,
starts to become very paranoid that she is being unfaithful to him. This started when the
girlfriend mentioned that a young man has joined the office where she works. Ever since then,
he checks the messages on her phone, follows her when she is with her friends and shows up at
her work place unannounced. He also demands to know where she is at all times, and they get
into arguments whenever she comes home from work late.
Q10: What do you think of the
man’s reaction?
1 = He has the right to do this
2 = I sympathise with him, but he should not do this
3 = He should never do this
S5’: Suppose that someone that you know, who has been dating her boyfriend for two years
now, starts to become very paranoid that he is being unfaithful to her. This started when the
boyfriend mentioned that a young woman has joined the office where he works. Ever since
then, she checks the messages on his phone, follows him when he is with his friends and shows
up at his work place unannounced. She also demands to know where he is at all times, and they
get into arguments whenever he comes home from work late.
Q11: What do you think of the
woman’s reaction?
1 = She has the right to do this
2 = I sympathise with her, but she should not do this
3 = She should never do this
READ: I am now going to read you a list of statements, and I would like you to tell me whether
you generally agree or disagree with the statement. There are no right or wrong answers.
1 = Fully
agree
2 =
Somewhat
agree
3 =
Somewhat
disagree
4 = Fully
disagree
98 =
Don’t
know/
Can’t
say
Q12: A good wife obeys her
husband even if she
disagrees with him
Q13: Family problems should only
be discussed with people in
the family
Q14: In order to bring up, raise or
educate a child properly, the
child sometimes needs to be
physically punished
310
Q15: I If a woman wears revealing
clothes, it is okay for men to
make some sexual comments
as she walks in her
neighbourhood
Q16: If a woman wears revealing
clothes, it is her fault if she is
raped
Q17: It is allowed for men to stare
at women they don’t know
Q18: In your opinion, does a man have a good reason to hit
his wife if: Yes No Don’t know
a) She does not complete her household work to his
satisfaction
1 0 98
b) She disobeys him 1 0 98
c) She refuses to have sex with him 1 0 98
d) She asks him whether he has other girlfriends 1 0 98
e) He suspects that she is unfaithful 1 0 98
f) He finds out that she has been unfaithful
g) She contracts a sexually transmitted disease
h) She goes out without telling him
i) She neglects the children
j) She argues with him
k) She burns the food
1
1
1
1
1
1
0
0
0
0
0
0
98
98
98
98
98
98
Q19: In your opinion, can a married woman refuse to have
sex with her husband if: Yes No Don’t know
a) She doesn't want to 1 0 98
b) He is drunk 1 0 98
c) She is sick 1 0 98
d) He mistreats her 1 0 98
e) She is menstruating
f) He does not want to use a condom/contraception
g) She finds out that he has girlfriends
1
1
1
0
0
0
98
98
98
4. Relationships and Reproductive Decisions
I want to ask a few questions about you and your current or most recent partner. By partner we
mean someone that you have a relationship with, for example a husband or wife, girlfriend or
boyfriend, lover or other romantic relationship.
Q1: Have you ever lived with a partner? 1 = Yes
0 = No >>>> Q3
Q2: How many partners have you ever lived with?
INCLUDE PARTNERS THE RESPONDENT IS
MARRIED TO AND THOSE YOU ARE
CURRENTLY LIVE WITH
_________________ times
Q3: Have you ever been married?
1 = Yes
0 = No >>>> Q5
311
THIS INCLUDES CURRENT MARRIAGE
Q4: How many times have you ever been married?
_________________ times
Q5: Are you currently in a relationship? 1 = Yes
0 = No >>>>> skip to Q52
Q6: Are you in a relationship with more than one
person?
1 = Yes
0 = No >>>>> skip to Q10
Q7: If yes, how many persons are you in a relationship
with?
THIS INCLUDES ALL RELATIONSHIPS, EVEN
THE PRIMARY RELATIONSHIP
__________________ number
Q8: Are you married to any of these other persons? 1 = Yes
0 = No
Q9: If yes, how many of these persons are you married
to?
THIS INCLUDES ALL MARRIAGES, EVEN THE
PRIMARY MARRIAGE
For the remainder of these questions I would like to ask you about your primary/main
partner. This could be the person that you live with.
Q10: Are you in a relationship with a man or a woman? 1 = Man
2 = Woman
Q11: How old were you when you entered this
relationship?
_________________ years
Q12: Was this an approximation 1 = Yes
0 = No
Q13: How did you get to know your
current partner?
1 = Arranged meeting
2 = We are family (e.g. cousin)
3 =
Friends/classmates/colleagues/neighbours/customer
4 = At party, festival or other social event
5 = Other
Q14: Who influenced the decision of
choosing your partner?
MULTIPLE ANSWERS
ALLOWED
1 = Myself
2 = Parents
3 = Extended family
4 = Church/Mosque/cult/sect
5 = Other
If respondent is man then skip to Q17
Q15: For women: Does your partner have any wives in
addition to your relationship?
THIS INCLUDES THE RESPONDENT IF S/HE IS
MARRIED
1 = Yes
0 = No >>>>> Q17
98 = Don’t know >>>>> Q17
Q16: For women: How many wives does your partner
have?
______________
98 = Don’t know
Q17: Does your partner have any partners/lovers in
addition to yourself?
1= Yes
0 = No >>>>> Q19
98 = Don’t know >>>>> Q19
Q18: How many partners/lovers does your partner have? ______________
98 = Don’t know
312
THIS INCLUDES THE RESPONDENT IF S/HE IS
IN A RELATIONSHIP
Q19: In general, do you and your current partner discuss
the following topics together:
Yes No Can’t say or
don’t want to
say
a) Things that have happened to your partner
during the day
1 0 3
b) Things that happen to you during the day 1 0 3
c) Your worries or feelings 1 0 3
d) your partner’s worries or feelings 1 0 3
Q20: In your relationship with
your current partner, how
often would you say that
you quarrelled?
1 = Less than once per month
2 = Once per month
3 = 2–3 times per month
4 = Once a week
5 = 2–3 times a week
6 = Almost every day
7 = Every day
1 = Yes, all
the time
2 = Yes,
sometimes
3 =
No,
never
Q21: In your relationship with your current recent partner,
has your partner ever gotten angry if you speak with
another man/woman?
Q22: In your relationship with your current partner, have
you ever gotten angry if your partner speaks with
another man/woman?
Q23: In your relationship with your current partner, does
your partner ever suspect that you are unfaithful, and
not trust you?
1 = Very often
2 = A few times
3 = Rarely
4 = Never
Q24: In your relationship with your current partner, do you
ever suspect that your partner is unfaithful, and not
trust him/her?
1 = Very often
2 = A few times
3 = Rarely
4 = Never
Q25: Are you currently living together? 1 = Yes >>>>> Q30
0 = No
Q26: Why not? 1 = I am not the first wife
2 = S/he is travelling for
work or education
3 = Too early in the
relationship
4 = Waiting to be married
5 = Need a bigger house
6 = In this part of Ghana, we don’t have to
live with our partner when we are married
7 = S/he is migrating for work or education
8 = I have migrated here for work or
education
9 = We have separated
10 = Other
If not, can I
ask you some
questions
about your
partner?
Q1: How old is
your partner?
Q2: What is the
highest level of
education that
your partner
has received?
1 = Pre-school
2 = Primary
3 = Middle/JSS/JHS
4 = Secondary/SSS/SHS
5 = TECH/VOC
6 = Higher
98 = Don’t know
Q3: What would
best describe
your partner’s
1 = Self-employed
2 = Employee
3 = Casual work
7 = Apprentice
8 = Retired/too old to
work
313
current work
situation?
4 = Not working and not
looking
5 = Not working and looking
for work
6 = Student
9 = Disabled/sick
10 = Too young to work
98 = Don’t know
Q4: Are you currently married? 1 = Yes
0 = No >>>>>>> Q50
Q5: How many years have you been
married?
Q6: Which type of marriage have you
had?
1 = Customary marriage
2 = Mohammedan/Islamic
3 = Ordinance
I would like to ask you a few questions about what would happen if you were to divorce. I know
this might not be happening to you now, but your opinion will help us better understand family
life in Ghana.
Q7: Who would be likely
to get custody of your
children (0–4 years)
in the case of a
divorce?
1 = Myself
2 = My spouse
3 = Other family
members
4 = Children
choose
5 = Divided in
half between
husband and
wife
6 = Girls will go to the mother and boys to the
father
7 = Divided depending on other assets
8 = Depends on who is at fault or who initiated
the divorce
66 = Other
98 = Don’t know
Q8: Who would be likely
to get custody of your
children (5–14 years)
in the case of a
divorce?
1 = Myself
2 = My spouse
3 = Other family
members
4 = Children
choose
5 = Divided in
half between
husband and
wife
6 = Girls will go to the mother and boys to the
father
7 = Divided depending on other assets
8 = Depends on who is at fault or who initiated
the divorce
66 = Other
98 = Don’t know
Q9: Would you expect the
main breadwinner to
provide financial
support for the
children in case of a
divorce?
1 = Yes
2 = Yes, but only if they
have custody of the
children
3 = Depends on who is at
fault or who initiated the
divorce
4 = Depends on the financial situation
0 = No
66 = Other
98 = Don’t know
Q10: Would you expect the
main breadwinner to
provide financial
support for his/her
1 = Yes
2 = Yes, but only if they
have custody of the
children
4 = Depends on the financial situation
0 = No
66 = Other
98 = Don’t know
314
spouse in case of a
divorce?
3 = Depends on who is at
fault or who initiated the
divorce
Q11: Who would be likely
to get immobile
assets, such as the
house and land, in
the case of a divorce?
1 = Myself
2 = My spouse
3 = Other family
members
4 = Divided in half
between husband
and wife
5 = Divided
depending on
custody of children
6 = Husband and wife get what belongs to
them
7 = Husband and wife get what they brought
into marriage
8 = Depends on who is at fault or who
initiated the divorce
66 = Other
98 = Don’t know
Q12: Who would be likely
to get mobile assets,
such as vehicles,
furniture, savings,
livestock, in the case
of a divorce?
1 = Myself
2 = My spouse
3 = Other family
members
4 = Divided in half
between husband
and wife
5 = Divided
depending on
custody of children
6 = Husband and wife get what belongs to
them
7 = Husband and wife get what they brought
into marriage
8 = Depends on who is at fault or who
initiated the divorce
66 = Other
98 = Don’t know
Q13: For men: Did your family pay a
bride price?
1 = Yes >>> Q41
0 = No
Q14: For men: Why not?
Then next module
1 = We could not
afford it
2 = We do not follow
this system
3 = We are still in the process of buying the
items
4 = Other
Q15: For men: What did the bride
price/dowry include?
1 = Cash
2 = Livestock
3 = Land
4 = Cloth/clothes
5 = Drinks
6 = Jewellery
7 = Suitcase
8 = Furniture
9 = Bible or Quran
10 = Other gifts
Q16: For men: Did your partner
contribute to the bride price?
1 = Yes
2 = No >>>> Then next module
Q17: For men: What did your partner
contribute?
1 = Cash
2 = Livestock
3 = Land
4 = Cloth/clothes
5 = Drinks
6 = Jewellery
7 = Suitcase
8 = Furniture
9 = Bible or Quran
10 = Other gifts
Q18: For women: Did your family
receive a bride price on your
behalf?
1 = Yes >>> Q46
0 = No
Q19: For women: Why
not?
Then next module
1 = My partner’s family
could not afford it
3 = My partner’s family are still in the
process of buying it
4 = Other
315
2 = We do not follow this
system
Q20: For women: What did the bride
price/dowry include?
1 = Cash
2 = Livestock
3 = Land
4 = Cloth/clothes
5 = Drinks
6 = Jewellery
7 = Suitcase
8 = Furniture
9 = Bible or Quran
10 = Other gifts
Q21: For women: Did you contribute to
the bride price?
1 = Yes
2 = No >>>> Next module
Q22: For women: What did you
contribute?
Then next module
1 = Cash
2 = Livestock
3 = Land
4 = Cloth/clothes
5 = Drinks
6 = Jewellery
7 = Suitcase
8 = Furniture
9 = Bible or Quran
10 = Other gifts
Q23: For women whose husband has
more than one wife: Are you the
first, second, third wife?
RANK _______
Q24: Are you engaged? 1 = Yes
0 = No
Q25: Are you in the process of a
customary marriage?
1 = Yes
0 = No
For those who are currently not in a relationship:
Q1: If you are not currently in a relationship,
have you had a relationship in the past?
IF RESPONDENT HAS HAD MORE
THAN ONE RELATIONSHIPS IN THE
PAST, PLEASE REFER TO THE MOST
RECENT ONE
1 = Yes, but we are now separated/broke up
2 = Yes, but s/he left/disappeared
3 = Yes, but we are now divorced
4 = Yes, but I am now widowed
0 = No >>>>>> NEXT MODULE
Q2: If 4 to Q52, were you ever asked to marry
the brother/sister of your spouse who
passed away?
1 = Yes
0 = No
Q3: If 1 to Q53, did you agree to marry a
brother /sister of your spouse who passed
away?
1 = Yes
0 = No
Q4: After your spouse passed
away, were your spouse's
assets (land, property,
money…) passed on to you?
1 = Yes, it was all passed on to
me
2 = Yes, some of it was passed
on to me
3 = No, it was given to our
children
4 = No, it was given to
my spouse’s family
5 = No, it was given to
someone else
I would like to ask you some questions about your most recent relationship.
Q5: Were you in a relationship with more
than one person?
1 = Yes
0 = No >>>>>> Q60
Q6: If yes, how many persons were you in a
relationship with?
__________________ number
316
THIS INCLUDES ALL RELATIONSHIPS,
INCLUDING THE PRIMARY
RELATIONSHIP
Q7: Were you married to any of these other
persons?
1 = Yes
0 = No >>>>>> Q60
Q8: If yes, how many of these persons were
you married to?
THIS INCLUDES ALL MARRIAGES,
INCLUDING THE PRIMARY
MARRIAGE
__________________ number
For the remainder of these questions I would like to ask you about your most recent
primary/main partner. This could be the person that you live with.
Q9: Were you in a relationship with a man or
a woman?
1 = Man
2 = Woman
Q10: How old were you when you entered this
relationship?
___________ years
Q11: Was this an approximation? 1 = Yes
0 = No
Q12: How old were you when you ended this
relationship?
____________ years
Q13: Was this an approximation 1 = Yes
0 = No
I would like to ask you some questions about your most recent partner.
Q14: How old was your partner when you
started your relationship?
Q15: How old was your partner when you
ended your relationship?
Q16: What was the highest level of
education that your partner
had received?
1 = Pre-school
2 = Primary
3 = Middle/JSS/JHS
4 = Secondary/SSS/SHS
5 = TECH/VOC
6 = Higher
98 = Don’t know
Q17: What would best describe
your most recent partner’s
work situation at the time?
1 = Self-employed
2 = Employee
3 = Casual work
4 = Not working and not
looking
5 = Not working and
looking for work
6 = Student
7 = Apprentice
8 = Retired/too old to work
9 = Disabled/sick
10 =Too young to work
98 = Don’t know
Q18: How did you get to know your most
recent partner?
1 = Arranged meeting
2 = We are family (e.g. cousins)
3 =
Friends/classmates/colleagues/neighbours/cus
tomer
4 = At party, festival or other social event
5 = Other
Q19: Who influenced the decision
of choosing your partner?
1 = Myself
2 = Parents
3 = Extended family
4 = Church/Mosque/cult/sect
5 = Other
If respondent is man skip to Q 73
317
MULTIPLE ANSWERS
ALLOWED
Q20: For Women: Did your partner have any wives
in addition to your relationship?
1 = Yes
0 = No >>>>> Q73
98 = Don’t know >>>>> Q73
Q21: For Women: How many wives did your
partner have?
THIS INCLUDES ALL MARRIAGES, EVEN
THE MARRIAGE WITH THE RESPONDENT
______________
98 = Don’t know
Q22: Did your partner have any partners/lovers in
addition to yourself?
1= Yes
0 = No >>>>> Q75
98 = Don’t know >>>>> Q75
Q23: How many partners/lovers did your partner
have?
THIS INCLUDES ALL RELATIONSHIPS,
EVEN THE RELATIONSHIP WITH THE
RESPONDENT
______________
98 = Don’t know
Q24: In general, did you and your most recent partner discuss
the following topics together:
Yes No Can’t say or
don’t want to
say
a) Things that had happened to your partner during the
day
1 0 3
b) Things that had happened to you during the day 1 0 3
c) Your worries or feelings 1 0 3
d) Your partner’s worries or feelings 1 0 3
Q25: In your relationship with
your most recent partner,
how often would you say
that you quarrelled?
1 = Less than once per month (once
in a while)
2 = Once per month
3 = 2–3 times per month
4 = Once a week
5 = 2–3 times a week
6 = Almost every day
7 = Every day
1 = Yes, all
the time
2 = Yes,
sometimes
3 = No,
never
Q26: In your relationship with your most recent partner,
did your partner ever get angry if you spoke with
another man/woman?
Q27: In your relationship with your most recent partner,
did you ever get angry if your partner spoke with
another man/woman?
Q28: In your relationship with your most recent partner,
did your partner ever suspect that you were
unfaithful, and not trust you?
1 = Very often
2 = A few times
3 = Rarely
4 = Never
Q29: In your relationship with your most recent partner,
did you ever suspect that your partner was
unfaithful, and not trust him/her?
1 = Very often
2 = A few times
3 = Rarely
4 = Never
Q30: Were you living together with your most
recent partner?
1 = Yes >>>>> Q83
0 = No
318
Q31: Why
not?
1 = I was not the first wife
2 = S/he was travelling for work or
education
3 = Too early in the relationship
4 = We were waiting to be married
5 = Needed a bigger house
6 = In this part of Ghana, we don’t have
to live with our partner when we are
married
7 = S/he was migrating for work or
education
8 = I had migrated for work or education
9 = We had separated
10 = Other
Q32: Were you married to your most recent
partner?
1 = Yes
0 = No >>> Q103
Q33: How many years were you married for?
Q34: Which type of marriage did you have? 1 = Customary marriage
2 = Mohammedan/Islamic
3 = Ordinance
I would like to ask you a few questions about what would happen if you were to divorce or if
you have had a divorce. I know this might not have happened to you, but your opinion will help
us better understand family life in Ghana.
Q35: Who would be likely to
get custody (or who got
custody) of your children
(0–4 years) in the case of a
divorce?
1 = Myself
2 = My ex-spouse
3 = Other family
members
4 = Children choose
5 = Divided in half
between husband and
wife
6 = Girls will go to the mother and
boys to the father
7 = Divided depending on other
assets
8 = Depends on who is at fault or
who initiated the divorce
66 = Other
98 = Don’t know
Q36: Who would be likely to
get custody (or who got
custody) of your children
(5–14 years) in the case of
a divorce?
1 = Myself
2 = My ex-spouse
3 = Other family
members
4 = Children choose
5 = Divided in half
between husband and
wife
6 = Girls will go to the mother and
boys to the father
7 = Divided depending on other
assets
8 = Depends on who is at fault or
who initiated the divorce
66 = Other
98 = Don’t know
Q37: Would you expect the
main breadwinner to
provide financial support
for the children, or did the
main breadwinner
provide financial support
for the children, in case of
a divorce?
1 = Yes
2 = Yes, but only if they have
custody of the children
3 = Depends on who is at fault or
who initiated the divorce
4 = Depends on the
financial situation
0 = No
66 = Other
98 = Don’t know
319
Q38: Would you expect the
main breadwinner to
provide financial support
for his/her spouse, or did
the main breadwinner
provide financial support
for his/her spouse, in case
of a divorce?
1 = Yes
2 = Yes, but only if they have
custody of the children
3 = Depends on who is at fault or
who initiated the divorce
4 = Depends on the
financial situation
0 = No
66 = Other
98 = Don’t know
Q39: Who would be likely to
get (or who got) immobile
assets, such as the house
and land, in the case of a
divorce?
1 = Myself
2 = My ex-spouse
3 = Other family members
4 = Divided in half between
husband and wife
5 = Divided depending on
custody of children
6 = Husband and wife get what
belongs to them
7 = Husband and wife get what
they brought into marriage
8 = Depends on who is at fault
or who initiated the divorce
66 = Other
98 = Don’t know
Q40: Who would be likely to
get (or who got) mobile
assets, such as vehicles,
furniture, savings,
livestock in the case of a
divorce?
1 = Myself
2 = My ex-spouse
3 = Other family members
4 = Divided in half between
husband and wife
5 = Divided depending on
custody of children
6 = Husband and wife get what
belongs to them
7 = Husband and wife get what
they brought into marriage
8 = Depends on who is at fault
or who initiated the divorce
66 = Other
98 = Don’t know
Q41: For men: Did your family pay a bride price? 1 = Yes >>> Q94 0 = No
Q42: For men: Why not?
Then next module
1 = We could not afford it
2 = We do not follow this
system
3 = We were still in the process of
buying it
4 = Other
For men: What did
the bride
price/dowry
include?
1 = Cash
2 = Livestock
3 = Land
4 = Cloth/clothes
5 = Drinks
6 = Jewellery
7 = Suitcase
8 = Furniture
9 = Bible or Quran
10 = Other gifts
Q1: For men: Did your partner contribute to
the bride price?
1 = Yes
2 = No >>>> Next module
Q2: For men: What did your partner
contribute?
Then next module
1 = Cash
2 = Livestock
3 = Land
4 = Cloth/clothes
5 = Drinks
6 = Jewellery
7 = Suitcase
8 = Furniture
9 = Bible or Quran
10 = Other gifts
Q3: For women: Did your family receive a
bride price/dowry on your behalf?
1 = Yes >>> Q99
0 = No
Q4: For women: Why
not?
Then next module
1 = My partner’s family could
not afford it
2 = We do not follow this system
3 = My partner’s family were still
in the process of buying it
4 = Other
320
Q5: For women: What did the bride
price/dowry include?
1 = Cash
2 = Livestock
3 = Land
4 = Cloth/clothes
5 = Drinks
6 = Jewellery
7 = Suitcase
8 = Furniture
9 = Bible or Quran
10 = Other gifts
Q6: For women: Did you contribute to the
bride price?
1 = Yes
2 = No >>>> Next module
Q7: For women: What did you contribute?
Then next module
1 = Cash
2 = Livestock
3 = Land
4 = Cloth/clothes
5 = Drinks
6 = Jewellery
7 = Suitcase
8 = Furniture
9 = Bible or Quran
10 = Other gifts
Q8: For women whose husband has more
than one wife: Were you the first, second,
third wife?
RANK _______
Q9: Were you engaged to your most recent
partner?
1 = Yes
0 = No
Q10: Were you in the process of a customary
marriage with your most recent partner?
1 = Yes
0 = No
321
Reproductive decisions
I would like to ask you some questions about your sexual health and life. Please remember that
all of this information will be kept confidential, and your name will be stored separately to the
rest of the information that we gather.
Q1: How old were you when you first had
sexual intercourse? _____________
AGE
99 = no answer
88 = no sexual experience
>>>> Q17
Q2: Did you give consent to your first
sexual partner?
1 = Yes
0 = No
Is the respondent in a relationship?
ENUMERATOR TO CHECK
RELATIONSHIP STATUS
1= Yes
0 = No >>>>> Q19
For those in a relationship: Now we would like to ask you a few questions regarding sexual and
reproductive health.
Q3: For women: Are you pregnant at the moment? 1 = Yes
0 = No >>>>> Q12
Q4: Before you were pregnant did you and your partner
use birth control (methods to delay or avoid
pregnancy?)
1 = Yes
0 = No >>>>> Q7
Q5: Who made this decision?
1 = Yourself
2 = Your partner
3 = Self and partner
4 = Someone else
Q6: What type of birth
control did you and
your partner use?
(MULTIPLE ANSWER
ALLOWED)
FOR ALL, GO TO Q12
AFTER
1 = Condom
2 = Female condom
3 = Hormonal pill
4 = Hormonal implant
5 = IUD (coil)
6 = Sterilisation
7 = Abstinence
8 = Withdrawal
9 = LAM
10 = Foam/jelly
11 = Diaphragm
12 = Rhythm
13 = Emergency contraception
14 = Injectable
15 = Other
Q7: Why did you and your
partner not use birth
control?
(MULTIPLE ANSWER
ALLOWED)
FOR ALL, GO TO Q12
AFTER
1 = Partner didn’t allow
2 = Someone else did not
allow
3 = Makes me feel weak/ill
or other side effects
4 = Makes sex less
pleasurable
5 = Didn’t feel the need to
6 = Religious beliefs
7 = It is expensive
8 = I felt embarrassed
9 = It is not available
10 = We are trying to have
children
11 = Me/partner post-
menopausal
12 = Homosexual relationship
13 = Me/partner is infertile
14 = Tubes are tied,
reproductive organs removed
or vasectomy
15 = We had not heard of
contraceptive methods
16 = We don’t have sex
17 = I was not menstruating
18 = Other
Q8: Do you and your partner use birth control (methods
to delay or avoid pregnancy?)
1 = Yes
0 = No >>>>> Q11
322
Q9: Who made this decision?
1 = Yourself
2 = Your partner
3 = Self and partner
4 = Someone else
Q10: What type of birth
control do you and your
partner use?
(MULTIPLE ANSWER
ALLOWED)
FOR ALL, GO TO Q12
AFTER
1 = Condom
2 = Female condom
3 = Hormonal pill
4 = Hormonal implant
5 = IUD (coil)
6 = Sterilisation
7 = Abstinence
8 = Withdrawal
9 = LAM
10 = Foam/jelly
11 = Diaphragm
12 = Rhythm
13 = Emergency contraception
14 = Injectable
15 = Other
Q11: Why are you and your
partner not using birth
control?
(MULTIPLE ANSWER
ALLOWED)
1 = Partner didn’t allow
2 = Someone else did not
allow
3 = Makes me feel
weak/ill or other side
effects
4 = Makes sex less
pleasurable
5 = Didn’t feel the need to
6 = Religious beliefs
7 = It is expensive
8 = I felt embarrassed
9 = It is not available
10 = We are trying to have
children
11 = Me/partner post-
menopausal
12 = Homosexual relationship
13 = Me/partner is infertile
14 = Tubes are tied,
reproductive organs removed
or vasectomy
15 = We have not heard of
contraceptive methods
16 = We don’t have sex
17 = I am not menstruating
18 = Other
Q12: Do you and your partner use a safe sex method to
avoid sexually transmitted diseases and infections?
1 = Yes
0 = No >>>>> Q14
Q13: Which method do you and your partner use?
(MULTIPLE ANSWER ALLOWED)
FOR ALL, GO TO Q15 AFTER
1 = Condom
2 = Female condom
3 = Abstinence
4 = Other
Q14: Why don’t you and your
partner use any method?
(MULTIPLE ANSWER
ALLOWED)
1 = Partner didn’t allow
2 = Someone else did not
allow
3 = Makes me feel
weak/ill or other side
effects
4 = Makes sex less
pleasurable
5 = Didn’t feel the need to
because we trust each
other
6 = Religious beliefs
7 = It is expensive
8 = I felt embarrassed
9 = It is not available
10 = We are trying to have
children
11 = We have not heard of safe
sex methods
12 = We don’t have sex
13 = Other
Q15: Over the last six months, or since you
met your partner, how often do you and
your partner engage in sexual
intercourse?
1 = Never
2 = Every day
3 = Several times a week
4 = Once a week
5 = Several times a month
323
6 = Occasionally
Q16: Do you think this is adequate for you? 1 = Yes
2 = No, it should be more
3 = No, it should be less
Q17: Has your partner ever refused to have
sexual intercourse with you for long
periods of time?
1 = Yes
0 = No
Q18: Who makes decisions about how many
children to have?
1 = Yourself
2 = Your partner
3 = Self and partner
4 = Someone else
5 = We have not discussed this (it is too
early)
6 = We cannot have children
Q19: Have you ever heard of female
circumcision?
1 = Yes >>>> if woman skip to q 21. If man
skip to next module
0 = No
Q20: In a number of countries, there is a
practice in which a girl may have part of
her genitals cut. Have you ever heard
about this practice?
1 = Yes >>>> If man skip to next module
0 = No >>>> Skip to next module
For women:
Q21: Have you yourself ever had your
genitals cut?
1 = Yes
0 = No >>>> Next module
Q22: Could you please tell me what was done
to you at this time.
Was any flesh removed from the genital
area?
1 = Yes
0 = No
98 = Don’t know
Was the genital area just nicked without
removing any flesh?
1 = Yes
0 = No
98 = Don’t know
Was your genital area sewn closed? 1 = Yes
0 = No
How old were you when this occurred? Age in completed years ____________
95 = During infancy
98 = Don’t know
Who cut (or nicked) the genitals? 1 = Traditional circumciser
2 = Traditional birth attendant
3 = Other traditional person
4 = Doctor
5 = Trained nurse/midwife
6 = Other health professional
7 = Other
98 = Don’t know
324
5. Sexual Violence
Please can you tell me if
anyone has ever done or
tried to do the following
things to you:
Clarify: Anyone includes your
partner, members of your
household, family, work
colleagues etc.
a) M
ade
inap
pro
pri
ate
sex
ual
com
men
ts t
o y
ou
th
at m
ade
yo
u f
eel
un
com
fort
able
b)
To
uch
ed y
ou
in
an
in
app
rop
riat
e
and
sex
ual
way
th
at m
ade
yo
u f
eel
un
com
fort
able
c) P
hy
sica
lly
fo
rced
yo
u t
o h
ave
sex
ual
inte
rco
urs
e o
r p
erfo
rm a
sex
ual
act
wh
en y
ou
did
no
t w
ant
to?
d)
Oth
erw
ise
forc
ed y
ou
to
hav
e se
xu
al
inte
rco
urs
e o
r p
erfo
rm a
sex
ual
act
wh
en y
ou
did
no
t w
ant
to?
Th
is c
ou
ld
be
by
bla
ckm
aili
ng
, th
reat
enin
g o
r
scar
ing
yo
u.
e) H
ad s
exu
al i
nte
rco
urs
e o
r o
ther
sex
ual
act
wit
ho
ut
yo
u b
ein
g a
ble
to
giv
e y
ou
r p
erm
issi
on
?
f) N
ot
use
d p
rote
ctio
n e
ven
aft
er y
ou
ask
ed?
g)
A s
exu
al p
artn
er d
id n
ot
rev
eal
to
yo
u t
hat
he/
she
had
HIV
(an
d t
hey
kn
ew a
bo
ut
it)?
h)
Had
sex
ual
in
terc
ou
rse
or
per
form
ed
a se
xu
al a
ct w
ith
so
meo
ne
bec
ause
yo
u
felt
lik
e y
ou
did
no
t h
ave
a ch
oic
e o
r
yo
u w
ere
wo
rrie
d a
bo
ut
the
reac
tio
n.
i) P
enet
rate
d y
ou
wit
h a
n o
bje
ct a
gai
nst
yo
ur
wil
l?
Q1: 1 = Yes
2 = Someone tried >>>
NEXT ITEM
0 = No >>> NEXT
ITEM
Q2: Did this happen when
you were younger
than 15 years old?
1 = Yes, 0 = No
Q3: Has this happened in
the last 12 months?
1 = Yes,
0 = No >>> NEXT
ITEM
325
Q4: In the last 12 months,
was the person who
did this to you from
your current
household? 1 = Yes all
of them, 2 = Some of
them, 0 = No >>Q6
Q5: Who did this to you
from your current
household? Use IDs
from Household roster
Then skip to Q8
Q6: Who did this to you?
Use codes above
Q7: What was the sex of
this person?
1 = Male, 2 = Female
Q8: I am sorry to ask
details about this
experience, but could
you tell me how they
physically forced you?
Q9: Over the last 12
months, how often has
this happened? Use
codes above
Q10: Did this result in any
injuries?
1 = Yes, 0 = No >>>>>
Q12
326
Q11: What type of injury?
Q12: How has this affected
your daily activities?
USE CODES BELOW
Q13: If 6 to Q12: How many
days were you not able
to work over the last 12
months?
Q14: If 5 to Q12: How many
days were you not able
to go to school over the
last 12 months?
Q15: When you were a child, did you ever observe this kind of behaviour between your
parents or other members of your household?
1 = Yes, parents
2 = Yes, other members
0 = No
Q16: Think of 10 of your closest neighbours, how many of them do you know, or think,
experienced this kind of behaviour over the last 12 months? 98 = Don’t know
_______________ out of 10
CODES FOR Q5 and 36: WHO DID THIS TO YOU? AND WHO DID YOU DO THIS TO? CODES FOR Q9 and 33: HOW OFTEN HAS
THIS HAPPENED?
1 = Partner
2 = Former partner
3 = In-laws/partner’s family
4 = Parents
5 = Sons, daughters
6 = Brothers, sisters
7 = Uncles, aunts
8 = Grandparents
9 = Cousins, nieces, nephews, other
relatives
10 = Priest or religious leader
11 = Other relatives who do not live in this household
but in the community
12 = Other relatives who live outside this community
13 = Friend
14 = Teacher
15 = Employer
16 = Classmates
17 = Work colleagues
18 = Police
19 = Other people in the community
20 = Other people outside the community
1 = Once only
2 = Once in a while
3 = Less than once per month
4 = Once per month
5 = 2–3 times per month
6 = Once a week
7 = 2–3 times a
week
8 = Almost every
day
9 = Every day
327
CODES FOR Q11: WHAT TYPE OF INJURY CODES FOR Q12: HOW HAS THIS AFFECTED YOU?
(select multiple)
CODES FOR Q8: HOW
WERE YOU
PHYSICALLY FORCED?
1 = Non-genital injuries
(Cuts, punctures, bites,
Scratches, abrasion, bruises)
2 = Genital injuries (bruising,
inflammation, tenderness,
abrasions, or lacerations)
3 = Chronic pain
4 = Gastrointestinal
disorders
5 = Gynaecological
complications
6 = Migraines and other
frequent headaches
7 = Sexually transmitted
infections
8 = Cervical cancer
9 = Loss of pregnancy
10 = Premature birth
0 = Not at all
1 = Unable to concentrate at
work
2 = Unable to concentrate at
school
3 = Unable to concentrate at
domestic work
4 = Could not take care of the
children
5 = I had to miss work
6 = I could not go to
school
7 = I couldn’t do
domestic work
8 = Lost confidence in
own ability
9 = Left work
10 = Left school
11 = Live in constant
fear
12 = Other
1 = Pinning you down, so
that you could not get up
2 = Beating you up if you
tried to move
3 = Beating you until you
became unconscious
4 = Otherwise hurting you
so that you would not
move
328
Q17: Over the last 12 months, how many times
were you injured due to sexual violence?
1 = Once/twice
2 = 3–5 times
3 = Many (more than 5) times
Q18: Over the last 12 months, were you ever hurt
badly enough due to sexual violence that
you needed health care?
IF YES: How many times?
1 = Times needed health care:
________
2 = Yes, but don’t know
3 = Not needed >>> NEXT MODULE
Q19: Did anyone stop you from trying to receive
health care for these injuries?
1 = Yes
0 = No >>>>> Q24
Q20: Was the person who stopped you from your
household?
1 = Yes
0 = No >>>>> Q22
Q21: Who stopped you from your household? Use Household IDs
Q22: Who stopped you from outside your
household?
Use perpetrator codes
Q23: How many times has someone stopped you
from receiving health care over the last 12
months?
__________________
Q24: Over the last 12 months, have you ever
received health care these injuries?
IF YES: All of the time, or sometimes?
1 = Times needed health care:
________
2 = Yes, but don’t know
3 = Not needed >>> NEXT MODULE
Q25: How much did you spend on health care for
your injuries including medicines, fees and
transport (IF MULTIPLE VISITS: during
your last health visit)?
Q26: For your injury, did you have to spend any
nights in any health facility over the last 12
months?
IF YES: How many nights (in total)?
Number of nights in health facility:
________
If none, enter ‘00’
Q27: During your last health care visit, did you
tell a health worker the real cause of your
injury?
1 = Yes >>> NEXT MODULE
0 = No
Q28: During your last health care visit, why did
you decide not to tell the real cause of your
injury?
MULTIPLE ANSWERS ALLOWED
1 = I was afraid they would tell someone
2 = I was afraid of repercussions from the
person who hurt me
3 = I was afraid they would not believe me
4 = I was afraid they would not care
5 = I was afraid they would treat me badly
6 = Some of the people who work there
know me or my family
7 = Other
329
If respondent has reported no sexual violence, Read out loud: I know that these questions
are very personal and some people would rather not reveal this sensitive information to a
stranger. So we are giving everyone, no matter what they said before, the chance to reveal
this information privately and discreetly without me watching you. No matter what you
answered before, I am going to give you the questionnaire, and let you choose your
responses without me looking. You can turn the page over as soon as you are done
Q29: Read to respondent: Before you were 15 years old, had
anyone ever touched you sexually or made you do
something sexual that you did not want to do?
Now hand questionnaire to respondent.
Q30: Now I would like you to answer the same question, but
when you were older than 15 years old.
Now hand questionnaire to respondent.
330
If respondent has been injured (Yes to ANY IN Q10) READ: I would like to learn more about the injuries resulting from the sexual violence you
have experienced
Please can you tell me if you have
ever done or tried to do the
following things to someone else:
Clarify: Anyone includes your
partner, members of your family,
work colleagues etc.
a) Made
inappropri
ate sexual
comments
to someone
that made
them feel
uncomfort
able
b) Touched
someone in
an
inappropri
ate and
sexual way
that made
them feel
uncomfort
able
c)
Physically
forced
someone to
have
sexual
intercourse
or perform
a sexual act
when they
did not
want to?
d)
Otherwise
forced
someone to
have
sexual
intercourse
or perform
a sexual
act? This
could be
by
blackmaili
ng,
threatening
or scaring
them
e) Had
sexual
intercourse
or
performed
a sexual act
with
someone
who was
not able to
give their
permission
?
f) Not used
protection
even after
someone
asked?
g) Not
reveal to a
sexual
partner
that you
had HIV
(and you
knew
about it)?
h) Felt that
someone
had sexual
intercourse
or
performed
a sexual act
with you
because
you felt like
they did not
have a
choice or
were
worried
about your
reaction
i)
Penetrated
someone
with an
object
against
their will?
Q1: 1 = Yes
2 = I tried >>> NEXT ITEM
0 = No >>> NEXT ITEM
Q2: Has this happened in the
last 12 months? 1 = Yes,
0 = No >>> NEXT ITEM
Q3: Over the last 12 months,
how often has this
happened? Above
Q4: Over the last 12 months,
was the person you did this
331
to from your household?
1 =Yes, 0 = No >>>>Q36
Q5: Over the last 12 months,
who did you do this to
from your household? Use
Household IDs
Q6: Over the last 12 months
who did you do this to?
Use codes above
Q7: What was the sex of this
person?
1 = Male, 2 = Female
332
Please can you tell me if anyone
has ever done, or tried to do, the
following things to you:
Clarify: Anyone includes your
partner, members of your
household, family, work colleagues
etc.
a) Slapped
you or
thrown
something
at you that
could hurt
you?
b) Pushed
you or
shoved
you?
c) Hit you
with
his/her fist
or with
something
else that
could hurt
you?
d) Kicked
you,
dragged
you or
beaten
you up?
e) Choked
or
strangled
you on
purpose?
f) Burnt
you on
purpose?
g) Used a
gun, knife
or other
weapon
against
you?
h) Poured
hazardous
chemicals
or
substance
s (e.g.
acid) on
you?
i) FOR
MEN
ONLY:
Kicked or
pulled
your
external
genitalia?
Q1: 1 = Yes
2 = Someone tried >>>
NEXT ITEM
0 = No >>> NEXT ITEM
Q2: Did this happen when you
were younger than 15
years old?
1 = Yes, 0 = No
Q3: Has this happened in the
last 12 months? 1 = Yes,
0 = No >>> NEXT ITEM
Q4: In the last 12 months, was
the person who did this to
you from your current
household? 1 = Yes all of
them, 2 = Some of them,
0 = No >>Q6
Q5: In the last 12 months, who
did this to you from your
current household? Use
IDs from Household roster
Then skip to Q8
333
Q6: In the last 12 months, who
did this to you?
Use codes below
Q7: What was the sex of this
person?
1 = Male, 2 = Female
Q8: Over the last 12 months,
how often has this
happened?
Use codes below
Q9: Over the last 12 months
did this result in any
injuries?
1 = Yes, 0 = No >>>>> Q11
Q10: What type of injury?
Q11: Over the last 12 months,
did this ever end in a fight
where you also hit back
and retaliated?
Clarify: to hurt the person
doing this to you? USE
CODES BELOW
Q12: How has this affected your
daily activities over the
last 12 months? USE
CODES BELOW
Q13: If 6 to Q12: How many
days were you not able to
work over the last 12
months?
334
Q14: If 5 to Q12: How many
days were you not able to
go to school over the last
12 months?
Q15: For women, if answered yes to any of the above: During the times you were hit,
were you ever pregnant?
1 = Yes, 0 = No >>> Q17
Q16: For women, if answered yes to any of the above: If yes, did something happen
to your pregnancy?
1 = Yes, I had a miscarriage
2 = Still birth
3 = Premature delivery
4 = Neurological
complications to foetus
5 = Neurological
complications to foetus
which include long term
disability
6 = Other options
0 = Nothing
Q17: When you were a child, did you ever observe this kind of behaviour between
your parents or other members of your household?
1 = Yes, parents
2 = Yes, other
members
0 = No
Q18: Think of 10 of your closest neighbours, how many of them do you know, or
think, experienced this kind of behaviour over the last 12 months?
98 = Don’t know
_______________ out of 10
CODES FOR Q6 and 37: WHO DID THIS TO YOU? AND WHO DID YOU DO THIS TO? CODES FOR Q8 and 34: HOW OFTEN
HAS THIS HAPPENED?
1 = Partner
2 = Former partner
3 = In-laws/partner’s family
4 = Parents
5 = Sons, daughters
6 = Brothers, sisters
7 = Uncles, aunts
8 = Grandparents
9 = Cousins, nieces, nephews,
other relatives
10 = Priest or religious leader
11 = Other relatives who do not live in this household but in the
community
12 = Other relatives who live outside this community
13 = Friend
14 = Teacher
15 = Employer
16 = Classmates
17 = Work colleagues
18 = Police
19 = Other people in the community
20 = Other people outside the community
1 = Once only
2 = Once in a
while
3 = Less than
once per month
4 = Once per
month
5 = 2–3 times per
month
6 = Once a week
7 = 2–3 times a week
8 = Almost every day
9 = Every day
335
CODES FOR Q10: WHAT TYPE OF
INJURY
CODES FOR Q11: DID YOU
EVER FIGHT BACK?
CODES FOR Q12: HOW HAS THIS AFFECTED YOU? (mark all
that apply)
1 = Cuts, punctures,
bites
2 = Scratch,
abrasion, bruises
3 = Sprains,
dislocations
4 = Burns
5 = Penetrating
injury, deep cuts,
gashes
6 = Ear or eye injury
7 = Fractures, broken
bones
8 = Broken teeth
9 = Other
1 = Never
2 = Once or twice
3 = Several times
4 = Many times/most of the time
0 = Not at all
1 = Unable to concentrate at work
2 = Unable to concentrate at school
3 = Unable to concentrate at
domestic work
4 = Could not take care of the
children
5 = I had to miss work
6 = I could not go to school
7 = I couldn’t do domestic
work
8 = Lost confidence in own
ability
9 = Left work
10 = Left school
11 = Live in constant fear
12 = Other
336
If respondent has been injured (Yes to ANY IN Q9) READ: I would like to learn more about the
injuries resulting from the physical violence you have experienced.
Q19: Over the last 12 months, how many
times were you injured due to
physical violence?
1 = Once/twice
2 = 3–5 times
3 = Many (more than 5) times
Q20: Over the last 12 months, were you
ever hurt badly enough due to
physical violence that you needed
health care?
IF YES: How many times?
1 = Times needed health care: ________
2 = Yes, but don’t know
3 = Not needed >>> NEXT MODULE
Q21: Did anyone stop you from trying to
receive health care for these injuries?
1 = Yes
0 = No >>>>> Q26
Q22: Was the person who stopped you
from your household?
1 = Yes
0 = No >>>>> Q24
Q23: Who stopped you from your
household?
Use Household IDs
Q24: Who stopped you from outside your
household?
Use perpetrator codes
Q25: How many times has someone
stopped you from receiving health
care over the last 12 months?
__________________
Q26: Over the last 12 months, have you
ever received health care these
injuries?
IF YES: All of the time, or
sometimes?
1 = Times needed health care: ________
2 = Yes, but don’t know
3 = Not needed >>> NEXT MODULE
Q27: How much did you spend on health
care for your injuries including
medicines, fees and transport (IF
MULTIPLE VISITS: during your last
health visit)?
Q28: For your injury, did you have to
spend any nights in any health
facility over the last 12 months?
IF YES: How many nights (in total)?
Number of nights in health facility:
________
If none, enter ‘00’
Q29: During your last health care visit,
did you tell a health worker the real
cause of your injury?
1 = Yes >>> NEXT MODULE
0 = No
Q30: During your last health care visit,
why did you decide not to tell the
real cause of your injury?
MULTIPLE ANSWERS ALLOWED
1 = I was afraid they would tell someone
2 = I was afraid of repercussions from the person
who hurt me
3 = I was afraid they would not believe me
4 = I was afraid they would not care
5 = I was afraid they would treat me badly
6 = Some of the people who work there know me or
my family
7 = Other
337
If respondent has reported no physical violence, Read out loud: I know that these
questions are very personal and some people would rather not reveal this sensitive
information to a stranger. So we are giving everyone, no matter what they said before,
the chance to reveal this information privately and discreetly without me watching you.
No matter what you answered before, I am going to give you the questionnaire, and let
you choose your responses without me looking. You can turn the page over as soon as
you are done
Q31: Read to respondent: Has anyone ever hurt you
physically by doing any of the things that I have just
mentioned?
Now hand questionnaire to respondent.
338
6. Physical Violence
Please can you tell me if you have
ever done or tried to do the
following things to someone else:
Clarify: Anyone includes your
partner, members of your family,
work colleagues etc.
a)
Slapped
or thrown
somethin
g at
someone
that could
hurt
them?
b) Pushed
or shoved
someone?
c) Hit
someone
with your
fist or
with
somethin
g that
could hurt
them?
d) Kicked,
dragged
or beaten
someone
up?
e) Choked
or
strangled
someone
on
purpose?
f) Burnt
someone
on
purpose?
g) Used a
gun, knife
or other
weapon
against
someone?
h) Poured
hazardous
chemicals
or
substance
s (e.g.
acid) on
someone?
i) Kicked
or pulled
someone’s
external
genitalia?
Q1: 1 = Yes
2 = I tried >>> NEXT ITEM
0 = No >>> NEXT ITEM
Q2: Has this happened in the
last 12 months? 1 = Yes,
0 = No >>> NEXT ITEM
Q3: Over the last 12 months,
how often has this
happened? Use codes above
Q4: Over the last 12 months, was
the person you did this to
from your household?
1 =Yes, 0 = No >>>>Q37
Q5: Over the last 12 months,
who did you do this to from
your household? Use
Household IDs
Q6: Over the last 12 months who
did you do this to?
Use codes above
339
Q7: What was the sex of this
person?
1 = Male, 2 = Female
7. Agency and Life Control
Q1: Please imagine a nine-step ladder, where on the
bottom, the first step, are those who are totally unable
to change their lives, while on step 9, the highest step,
stand those who have full control over their own life.
On which step are you?
__________________
How much do you agree with the following
statements?
1 = Strongly
agree
2 = Somewhat
agree
3 =
Neutral
4 = Somewhat
disagree
5 = Strongly
disagree
Q2: In uncertain times, I usually expect the best
Q3: It’s easy for me to relax
Q4: If something can go wrong, it will
Q5: I’m always optimistic about my future
Q6: I enjoy my friends a lot
Q7: It’s important for me to keep busy
Q8: I hardly ever expect things to go my way
Q9: I don’t get upset too easily
Q10: I rarely count on good things happening to me
Q11: Overall, I expect more good things to happen to me
than bad
340
Please can you tell me if anyone
has ever (even when you were a
child) done, or tried to do, the
following things to you:
Clarify: Anyone includes your
partner, members of your household,
family, work colleagues etc.
a) Insulted,
belittled, or
humiliated you in
private or in front
of other people
b) Threatened to
abandon you,
ignored you or
treated you
indifferently
c) Done things to scare
or intimidate you on
purpose? (For example
by breaking something
in front of you)
d) Threatened to
use a gun, knife or
other weapon
against you?
e) Threatened to
hurt you or
someone you
care about with
something other
than a weapon?
This includes
threats to take
away children or
someone you
care about
Q1: 1 = Yes
2 = Someone tried >>> NEXT ITEM
0 = No >>> NEXT ITEM
Q2: Did this happen to you when you
were younger than 15?
1 = Yes, 0 = No
Q3: Has this happened in the last 12
months?
1 = Yes, 0 = No >>> NEXT ITEM
Q4: Did this behaviour make you feel
worthless or make you doubt
yourself?
1 = Yes, all the time
2 = Yes, sometimes
3 = No, not at all
Q5: In the last 12 months, was the
person who did this to you from
your current household? 1 = Yes all
of them, 2 = Some of them, 0 = No
>>Q7
341
Q6: In the last 12 months, who did this
to you from your current
household? Use IDs from
Household roster
Then skip to Q9
Q7: In the last 12 months, who did this
to you? Use codes above
Q8: What was the sex of this person?
1 = Male, 2 = Female
Q9: Over the last 12 months, how often
has this happened?
Use codes above
Q10: When you were a child, did you ever observe this kind of behaviour between
your parents or other members of your household?
1 = Yes, parents
2 = Yes, other members
0 = No
Q11: Think of 10 of your closest neighbours, how many of them do you know, or
think, experienced this kind of behaviour over the last 12 months?
98 = Don’t know
_______________ out of 10
CODES FOR Q6 and 18: WHO DID THIS TO YOU? AND WHO DID YOU DO THIS TO? CODES FOR Q9 and 15: HOW OFTEN HAS
THIS HAPPENED?
1 = Partner
2 = Former partner
3 = In-laws/partner’s
family
4 = Parents
5 = Sons, daughters
6 = Brothers, sisters
7 = Uncles, aunts
8 = Grandparents
9 = Cousins, nieces, nephews,
other relatives
10 = Priest or religious leader
11 = Other relatives who do not
live in this household but in the
community
12 = Other relatives who live
outside this community
13 = Friend
14 = Teacher
15 = Employer
16 = Classmates
17 = Work colleagues
18 = Police
19 = Other people in the community
20 = Other people outside the
community
1 = Once only
2 = Once in a while
3 = Less than once per
month
4 = Once per month
5 = 2–3 times per month
6 = Once a week
7 = 2–3 times a week
8 = Almost every
day
9 = Every day
342
Please can you tell me if you have
ever done, or tried to do, the
following things to someone else:
Clarify: Anyone includes your
partner, members of your family,
work colleagues etc.
a) Insulted,
belittled, or
humiliated
someone in
private or in front
of other people
b) Threatened to
abandon someone,
ignored someone or
treated them
indifferently
c) Done things to
scare or intimidate
someone on
purpose? (For
example by breaking
things in front of
someone)
d) Threatened to
use a gun, knife or
other weapon
against someone?
e) Threatened to
hurt someone or
someone they
care about with
something other
than a weapon?
This includes
threats to take
away children or
someone they
care about
Q1: 1 = Yes
2 = I tried >>> NEXT ITEM
0 = No >>> NEXT ITEM
Q2: Has this happened in the last 12
months?
1 = Yes, 0 = No >>> NEXT ITEM
Q3: Over the last 12 months, how
often has this happened?
Use codes above
Q4: Over the last 12 months, was the
person you did this to from your
household?
1 =Yes, 0 = No >>>>Q17
Q5: Over the last 12 months, who did
you do this to from your
household? Use Household IDs
Q6: Over the last 12 months who did
you do this to? Use codes above
Q7: What was the sex of this person?
1 = Male, 2 = Female
343
8. Economic Violence
Q1: Do you have a bank account? 1 = Yes, 0 = No >>>>>>Q3
Q2: How much savings do you have in your bank account? _________________
Q3: Do you use mobile banking? 1 = Yes, 0 = No >>>>>Q5
Q4: How much savings do you have in your mobile banking account? _________________
Q5: Do you have any other form of savings? 1 = Yes, 0 = No >>>>>Q7
Q6: How much are these total savings? __________________
Please can you tell me if anyone
has ever (even when you were a
child) done, or tried to do, the
following things to you:
Clarify: Anyone includes your
partner, members of your household,
family, work colleagues etc.
a) Refused to
give you
enough chop
money even
though you
think he/she
has enough
money to
spend on other
things?
b) Taken
cash or
withdrawn
money from
your bank
account or
other savings
without
permission?
c)
Controlled
your own
belonging
s and/or
your
spending
decisions?
d)
Destroyed
or damaged
property
that you
have
material
interest in?
e)
Prohibited
you from
working or
forced you
to quit your
work?
f) Forced you
to work
against your
will?
g) Refused to
give you or
denied you food
or other basic
needs?
Q7: 1 = Yes
2 = Someone tried >>> NEXT
ITEM
0 = No >>> NEXT ITEM
Q8: Has this happened in the last 12
months?
1 = Yes, 0 = No >>> NEXT ITEM
Q9: In the last 12 months, was the
person who did this to you from
your current household? 1 = Yes
all of them, 2 = Some of them,
0 = No >>Q11
344
Q10: In the last 12 months, who did
this to you from your current
household? Use IDs from
Household roster
Then skip to Q7
Q11: In the last 12 months, who did
this to you? Use codes below
Q12: What was the sex of this person?
1 = Male, 2 = Female
Q13: Over the last 12 months, how
often has this happened?
Use codes above
Q14: When you were a child, did you ever observe this kind of behaviour between
your parents or other members of your household?
1 = Yes, parents
2 = Yes, other members
0 = No
Q15: Think of 10 of your closest neighbours, how many of them do you know, or
think, experienced this kind of behaviour over the last 12 months?
98 = Don’t know
_______________ out of 10
CODES FOR Q11 and 21: WHO DID THIS TO YOU? AND WHO DID YOU DO THIS TO? CODES FOR Q13 and 18: HOW OFTEN HAS
THIS HAPPENED?
1 = Partner
2 = Former partner
3 = In-laws/partner’s
family
4 = Parents
5 = Sons, daughters
6 = Brothers, sisters
7 = Uncles, aunts
8 = Grandparents
9 = Cousins, nieces, nephews,
other relatives
10 = Priest or religious leader
11 = Other relatives who do not
live in this household but in
the community
12 = Other relatives who live
outside this community
13 = Friend
14 = Teacher
15 = Employer
16 = Classmates
17 = Work colleagues
18 = Police
19 = Other people in the community
20 = Other people outside the
community
1 = Once only
2 = Once in a while
3 = Less than once per
month
4 = Once per month
5 = 2–3 times per month
6 = Once a week
7 = 2–3 times a
week
8 = Almost every
day
9 = Every day
345
Please can you tell me if you
have ever done, or tried to do,
the following things to someone
else:
Clarify: Anyone includes your
partner, members of your family,
work colleagues etc.
a) Refused to
give someone
enough chop
money even
though you
have enough
money to
spend on
other things?
b) Taken
cash or
withdrawn
money from
someone’s
bank account
or other
savings
without
permission?
c)
Controlled
someone’s
belongings
and/or their
spending
decisions?
d) Destroyed
or damaged
property that
someone has
a material
interest in?
e) Prohibited
someone
from
working or
forced
someone to
quit their
work?
f) Forced
someone to
work against
their will?
g) Refused
to give or
denied
someone
food or other
basic needs?
Q16: 1 = Yes
2 = I tried >>> NEXT ITEM
0 = No >>> NEXT ITEM
Q17: Has this happened in the last 12
months?
1 = Yes, 0 = No >>> NEXT ITEM
Q18: Over the last 12 months, how
often has this happened? Use
codes above
Q19: Over the last 12 months, was the
person you did this to from your
household?
1 =Yes, 0 = No >>>>Q7
Q20: Over the last 12 months, who did
you do this to from your
household? Use Household IDs
Q21: Over the last 12 months who did
you do this to? Use codes above
Q22: What was the sex of this person?
1 = Male, 2 = Female
346
9. Leaving Home
Q1: Over the last 12 months, have you ever left your home, even if
only for the night, because of a problem or situation at home?
1 = Yes
0 = No >>> NEXT
MODULE
Q2: Was this following any of the treatment or behaviour we have
previously discussed?
1 = Yes
0 = No >>>>Q4
Can you tell me which
treatment or behaviour in
particular?
Mark all that apply
Produce list of violence and behaviour that respondent
has said have happened to him/her over last 12 months
Q3: What were the reasons you left
the last time?
MARK ALL MENTIONED
0 = No particular reason
1 = Encouraged by friends/family
2 = Could not endure more
3 = Badly injured
4 = Afraid person/persons would kill me
5 = The person/persons who did this to me threatened or
tried to kill me
6 = The person/persons who did this to me threatened or
hit the children
7 = Saw that children are suffering
8 = Thrown out of the home
9 = Encouraged by organisation
10 = Encouraged by health workers/teachers
11 = Encouraged by police
12 = Other
Q4: How long did you stay away
the last time?
RECORD NUMBER OF
DAYS OR MONTHS
1 = Number of days (if less than one month)
_____________
2 = Number of months (if 1 month or more)
_____________
3 = Left or did not return to person who did this to me
>>>> Next module
Q5: Why did you return?
MARK ALL MENTIONED
1 = Didn’t want to leave children
2 = Sanctity of marriage
3 = For sake of family/children
4 = Couldn’t financially support children
5 = Loved person who did this to me
6 = Person who did this to me asked me to come back
7 = Encouraged to return by family
8 = Encouraged to return by health workers/teachers
9 = Encouraged to return by police
10 = Forgave person who did this to me
11 = Thought person who did this to me would change
12 = Person who did this to me threatened
partner/children
13 = Ran out of money
14 = Do not have another home to go to
15 = I was concerned about safety of people I was with
16 = I was concerned about safety of people left behind
347
17 = Other
10. Access to Services
Verify if respondent has ever experienced any violence? 1 = Yes
0 = No >>>>> Q8
Q1: After experiencing any of the types of violence
and treatment discussed previously, have you
ever tried to approach a person or organisation
for help or support?
1 = Yes >>>>>>>Q3
0 = No
Q2: Why not?
THEN GO TO
Q12
1 = Did not know who to ask
2 = Person/organisation too far
3 = Nobody can help me
4 = I don’t need help
5 = If my partner finds out it will
cause trouble
6 = If other family member finds out
it will cause trouble
7 = I don’t trust anyone
8 = Violence normal/not
serious
9 = Embarrassed/ashamed
10 = Afraid would not be
believed
11 = Afraid would be blamed
12 = Afraid would end
relationship
13 = Afraid would lose
children
14 = Bring bad name to family
15 = Other
Q3: What made you
approach
someone?
1 = Encouraged by friends/family
2 = Could not endure more
3 = Badly injured
4 = Afraid partner would kill me
5 = Partner threatened or tried to kill
me
6 = Partner threatened or hit
the children
7 = Saw that children are
suffering
8 = Thrown out of the home
9 = Encouraged by
organisation
10 = Other
Q4: Q5: Q6: Q7:
Who did you approach? How satisfied were
you with the way
that they dealt with
your case?
1 = Completely
satisfied
2 = Somewhat
satisfied
3 = Somewhat
dissatisfied
4 = Completely
dissatisfied
Did you have
to pay
anything to
this person or
organisation?
1 = Yes, I paid
a fee
2 = Yes, I paid
a bribe
0 = No >>>>
Next Item
How much
did you have
to pay?
A A family member or friend
B Health centre/hospital/other
health service provider
C Police
D Domestic Violence and Victim
Support Unit (DOVVSU)
348
E Shelter
F NGO, CSO or social worker
G Lawyer or member of the Court
H Legal aid group
I Queen mother of your
community
J Traditional leaders
K Religious leaders
L Religious group
M Community group
N Other
Thank you very much for answering these questions. We are almost at the end, and we have only
three more topics we would like to briefly discuss. First, we would like to ask you about access to
services in your area.
Q8: How long does it take you to get to your
nearest police station?
________ minutes
98 = Don’t know
Circle one:
1 = Foot
2 = Car
3 = Bicycle
4 = Motorcycle
5 = Tricycle
6 = Boat/canoe
7 = Public transport (e.g. bus or tro-tro)
8 = Truck
9 = Animal drawn cart
10 = Tractor
Q9: Does your nearest police station have a
domestic violence victim support unit
(DOVVSU)?
1 = Yes >>>>Q11
0 = No
98 = Don’t know
Q10: Do you know where the closest domestic
violence victim support unit (DOVVSU) is?
1=Yes
0=No
Q11: If not, how long would it take you to get to
the nearest domestic violence victims support
unit (DOVVSU)?
______________ minutes
98 = Don’t know
Circle one:
1 = Foot
2 = Car
3 = Bicycle
4 = Motorcycle
5 = Tricycle
6 = Boat/canoe
7 = Public transport (e.g. bus or tro-tro)
8 = Truck
9 = Animal drawn cart
10 = Tractor
349
How much would you trust the following if a friend
needed help after being physically assaulted in the home
by their partner?
1 = Completely trust
2 = Somewhat trust
3 = Somewhat distrust
4 = Completely distrust
98 = Never heard of them
Q12: Health centre/hospital/other health service provider
Q13: Police
Q14: Domestic Violence and Victim Support Unit (DOVVSU)
Q15: Shelter
Q16: NGO, CSO or social worker
Q17: Lawyer or member of the Court
Q18: Legal aid group
Q19: Queen mother of your community
Q20: Traditional leaders
Q21: Religious leaders
Q22: Religious group
Q23: Community group
Q24: Does Ghana have a law
against domestic violence?
1 = Yes
2 = Yes, but I do not
understand it
3 = Yes, but I do not know
what is in there exactly
4 = No >>>>>>NEXT
MODULE
98 = Don’t know
>>>>NEXT MODULE
Q25: If yes, where have you heard
about it?
1 = Friends/family
2 = Neighbours
3 = Community leaders
4 = Religious leaders
5 = Organisation
6 = Police
7 = School
8 = TV
9 = Radio
10 = Newspaper
11 = Books
12 = Other
11. Health
I would now like to ask you a few questions about your health and use of health services.
Q1: In general, would you describe your
overall health as …?
1 = Very good
2 = Good
3 = Fair
4 = Poor
5 = Very poor
Q2: Do you have any chronic conditions? 1 = Yes; 0 = No >>>>>Q4
Q3: Please indicate which
chronic condition(s)
you have?
MULTIPLE
ANSWERS
1 = Diabetes
2 = Asthma
3 = Lung disease
4 = Heart disease
5 = Arthritis or other
rheumatic disease
6 = Cancer
7 = High blood pressure
8 = Physical disability which restricts
certain activity
9 = Mental disability which restricts
certain activity
10 = Other chronic condition
Q4: Which of the following describes your
smoking habit?
1 = I used to smoke but not anymore
2 = I currently smoke >>> SKIP TO Q6
0 = I never smoked >>> SKIP TO Q7
1 = Daily 2 = Occasionally
Q5: Did you use to smoke daily or occasionally?
350
Q6: Do you smoke daily or occasionally?
Q7: Do you drink alcohol? 1 = Yes; 0 = No >>>>if in relationship Q9 if not Q11
Q8: How often do you drink
alcohol?
1 = Every day or nearly every day
2 = More than half the week but not
everyday
3 = Once or twice a week
4 = 1–3 times a month
5 = Less than once a
month
Q9: Does your partner drink alcohol? 1 = Yes, 0 = No >>>> Q11
Q10: How often does your
partner drink alcohol?
1 = Every day
2 = More than half the week but not
every day
3 = Once or twice a week
4 = 1–3 times a month
5 = Less than once a
month
Q11: Has anyone ever stopped you from or did not allow you to receive
health care when you needed it? Clarify: This could be by not allowing
respondent to visit health care facility or refusing to pay the fees or
costs of medicine.
1 = Yes
0 = No >>>>>
Q17
Q12: Has this happened over the last 12 months? 1 = Yes; 0 = No >>>> Q17
Q13: Was the person who stopped you from your household (over
the last 12 months)?
1 = Yes; 0 = No >>>>> Q15
Q14: Who stopped you from your household (over the last 12 months)? Use Household IDs
Q15: Who stopped
you from
outside your
household
(over the last
12 months)?
1 = Partner
2 = Former partner
3 = In-laws/partner’s family
4 = Parents
5 = Sons, daughters
6 = Uncles, aunts
7 = Grandparents
8 = Cousins, nieces, nephews,
other relatives
9 = Priest or religious leader
10 = Other relatives who do not live in this
household but in the community
11 = Other relatives who live outside this
community
12 = Friend
13 = Teacher
14 = Employer
15 = Work colleagues
16 = Police
17 = Other people in the community
18 = Other people outside the community
Q16: How many times has someone stopped you from receiving health care
over the last 12 months?
___________
Q17: In the last 30 days, have you felt sick or unwell? 1 = Yes; 0 = No >>>Q20
Q18: In the last 30 days, did you personally
consult a doctor or other professional
or health worker or traditional healer
because you were sick? Clarify: This
does not include health check-ups.
IF YES: Whom did you consult?
MULTIPLE ANSWERS ALLOWED
1 = No one
2 = Doctor
3 = Nurse
4 = Counsellor
5 = Traditional healer
6 = Pharmacist/chemical
seller
7 = Midwife
8 = Traditional birth
attendant
9 = Herbalist
10 = Spiritualist
11 = Prayer camp
12 = Other
Q19: Were you unable to work because of your sickness? IF YES: How many days? _________
During the last 30 days, about how
often did …
All the
time
Most
of the
time
Some
of the
time
A little
of the
time
None
of the
time
Q20: you feel nervous?
Q21: you feel hopeless?
Q22: you feel restless or fidgety?
Q23: you feel that everything was an effort?
351
Q24: you feel so depressed that nothing could
cheer you up?
Q25: you feel worthless?
Q26: has the thought of ending your life been
on your mind?
If there are children in the household aged 14 and younger
Adults use certain ways to teach children the right behaviour or to address a behaviour problem.
I will read various methods that are used and want you to tell me if you have used this method
with any children (aged 14 and younger) in your household in the past month.
In the last month, have you or
anyone else in your household
1 = Yes, 0 = No
Q1: Taking away privileges
Q2: Forbidding something that he/she liked
Q3: Not allow him/her to leave the house
Q4: Explaining why the behaviour was wrong
Q5: Giving him/her something else to do
Q6: Ignoring/refusing to communicate to the child
Q7: Shouting, yelling at or screaming at him/her
Q8: Calling him/her dumb, lazy or another name like that
Q9: Locking child up in room
Q10: Shaking him/her
Q11: Spanking, hitting, pushing or slapping him/her on the
bottom with bare hand
Q12: Hitting him/her on the bottom or elsewhere on the
body with something like a belt, comb, hairbrush, cane,
stick or other hard object
Q13: Hitting or slapping him/her on the face, head or ears
Q14: Hitting or slapping him/her on the hand, arm or leg
Q15: Beating him/her up, that is hitting him/her over and
over as hard as one could
Q16: Physically hurting them in any other way (e.g. burning)
12. Children
Finally, I would like to ask you a few questions about how the children of this household spend
their time.
Q17: How many hours do you spend with your
child/children (aged 14 and below) each day?
98 = Don’t know
99 = Not applicable
Q18: How many hours do you spend supervising your
child/children (aged 14 and below) during homework
each day? Should be less than Q17
98 = Don’t know
99 = Not applicable
Q19: How many hours does your child spend with other
adult household members?
98 = Don’t know
99 = Not applicable
Q20: Do you regularly attend Parent-Teacher Association
(PTA) meetings?
1 = Always
2 = Often
352
3 = Sometimes
4 = Rarely
5 = Never
99 = Not applicable
Q21: Does someone else from this household regularly
attend PTA meetings?
1 = Yes
0 = No
Thank you very much for your time! We know it is not easy to answer all these questions, and
we really appreciate that you took the time and patience. Since we have talked about issues
related to violence, we recommend that you do not speak to others about this part of the
questionnaire. Otherwise, some people in the community might get upset with you or others who
answered the survey. Are there any questions you have for us, or is there anything that you
would like to let us know?
_________________________________________________
353
Q3: Qualitative instruments
1. Focus Group Discussion Guideline – Domestic Violence Ghana 2015
Approach: For four separate and mixed FGDs with men and women. 1:30 hours /8–12
people
Group 1: Married/living together men (aged 18–50)
Group 2: Married /living together women (aged 18–50)
Group 3: Unmarried/youth (aged 18-25)
Group 4: Mixed (aged 50+)
Facilitator introduces the team, explains consent and the purpose
Thanks everyone for coming. Good morning/afternoon/evening and welcome. Thank you
all for taking the time to be part of this discussion. We very much appreciate your
willingness to participate.
My name is [name of facilitator], and I will serve as the moderator for today’s focus group
discussion. Assisting me is [name of assistant], who will take notes. This study is part of
the National Survey on Domestic Violence 2015 being conducted by the Institute of
Development Studies, University of Sussex (IDS) for the Ministry of Gender, Children
and Social Protection and DfiD, UK.
Participation in this research is voluntary. The discussion may last for about 45 to 90
minutes. All information provided will be kept secret. We will not ask your real name
and/or your address. You can withdraw from this interview anytime you want, or refuse
to answer any questions that you are not comfortable with.
The purpose of today’s discussion is to learn from you about:
The causes and consequences of domestic violence
Which services survivors of domestic violence use
What works well, what does not
How these services can be improved or strengthened
The findings will also inform civil society organisations and help them to reach out
to survivors of domestic violence and lobby for more programmatic interventions
and changes in laws.
There will be a report that will be published internationally.
We will be spending a few days in the community trying to find out about the different
roles of women and girls and men and boys in the community and in the family, and
discuss whether and how views and experiences with domestic violence has changed
compared to the past.
354
Facilitator explains the consent forms and the value of recording, and checks
permission.
Facilitator and Note-taker help participants complete consent forms.
Note-taker explains that we give people a number and use the numbers for the note-
taking.
Facilitator asks group for permission to record the discussion.
Rules
We would like you to feel safe here. We are here to learn from you; you have the knowledge
to help us understand these issues. We are not going to judge you and will listen to
everything you have to say. To do this, we would like to employ some guidelines for the
session to make sure we are safe and have a good discussion; the guidelines will apply to
all of us.
1. THERE ARE NO RIGHT OR WRONG ANSWERS
Every person’s experiences and opinions are important.
Speak up whether you agree or disagree.
It’s ok to have a different opinion; but everyone’s opinion matters equally.
We are interested to hear everybody’s opinion.
2. WE WOULD LIKE YOU TO DO THE TALKING.
Don’t feel like you have to respond to me all the time.
However, we would like everyone to participate.
So if you’re talking a lot, I may ask you to give others a chance.
And, I may call on you if I haven’t heard from you in a while.
3. WHAT IS SAID IN THIS ROOM STAYS HERE
Your experiences and input is very important and will feed into the research report – but
not what you say individually, unless you would like to be named.
We have name tents here in front of us today so that we can all know who’s sitting here
with us, but no names will be included in any reports!
Discussion starts.
Facilitators start with the questions. They do not read the headings.
Community’s perception of violence: What is violence?
1. Is it acceptable if a man beats his wife for disagreeing with him? (Allow different
opinions and allow participants to state reasons for their answers.)
If yes, why? …………………………………
If no, why? ……………………………
355
2. Is it acceptable if a woman beats her husband for having a child with someone
outside the family? (Allow different opinions and allow participants to state reasons
for their answers.)
If yes, why? …………………………………
If no, why? ……………………………
3. Is it acceptable if a man beats his wife for refusing to have sex with him? (Allow
different opinions and allow participants to state reasons for their answers.)
If yes, why? …………………………………
If no, why? ……………………………
4. Is it acceptable if a man refuses to give ‘chop money’ to his wife? (Allow different
opinions and allow participants to state reasons for their answers.)
If yes why? …………………………………
If no, why? ……………………………
5. Is it acceptable for men to tell women they cannot work?
If yes why? …………………………………
If no, why? ……………………………
6. Is it acceptable for men to beat children and youth in the home?
If yes why? …………………………………
If no, why? ……………………………
7. Is it acceptable for women to beat children and youth in the home?
If yes why? …………………………………
If no, why? ……………………………
8. Is it acceptable for fathers to stop their daughters from attending school?
If yes, why? ……………………………………………………..
If no, why? ………………………………………………………..
When should one intervene?
Allow different opinions and allow participants to state reasons for their answers.
9. What would you do if you were passing along a courtyard in your neighbourhood
and saw a man and woman shouting at each other?
10. What would you do if you saw the man now beating the woman?
11. What would you do if you saw a woman now beating the man?
12. What would you do if you saw the man/woman take out an object? What would
you do if you saw the woman/man now hurting the man/woman with the object?
Causes of domestic violence
13. Do we know whether there is disagreement and fights by people/men and women
from the same family in this community? If Yes, probe: [Who is involved in these
fights? What is the fighting about? How often do we hear or see these fights?]
14. So, why is there fighting within our families in this community
15. What are the various types of disagreement within families?
16. What typically happens during these fights/disagreements? (Describe these fights.)
356
17. Do men also experience violence in the family? What does this look like? How is
this different from how women experience violence in the domestic context?
18. Are weapons involved? (If yes, what type of weapons?)
19. Who is affected by these fights? How? Why? (Probe: children, elderly members,
others in family?)
20. What is the immediate impact? And what is the long-term impact on those affected?
Intervention and mediation
21. How do these domestic fights come to an end? Who do people go to for help? Who
is involved?
22. In your community, what kind of services are there to help people who are hurt by
their family? (Probe: list the services to understand the ‘in principle’ facets of
government support.)
23. Which of these services do people use the most? Why?
24. What has been your experience with government support for people hurt by other
people in their family? [Probe and discuss satisfaction with police handling of case
reporting/help at health centres.]
Social cost of domestic violence
25. Are others in the community affected by a family’s occasions of violence/blow-up?
Identify who. How are these persons affected? [Probe: different categories of
persons identified and how they are impacted by the violence.]
26. What are the costs to the rest of the community by a family’s experience of violence?
What would make violence acceptable to people/the community?
27. What would make members of the family turn a blind eye to violence within a
family?
28. What are the expectations of masculinity (breadwinner, disciplinarian) that can
make men feel they need to behave in a certain dominant or violent way?
29. What are the expectations of femininity (submissiveness) that can make women feel
they need to accept certain dominant or violent ways from men?
30. Discuss what are social pressures that can lead to people hurting one another in a
family or household – family and community pressure to encourage or discourage
prosecution.
31. What are the government pressures to encourage or discourage prosecution of
cases?
N.B. These are just guidelines. If any interesting additional issues emerge, the
discussion will also explore these. Facilitator asks participants if they have any
questions for us. Facilitator thanks participants for their time and the interesting
information they have shared.
357
2. Observation and In-Depth Interview Guideline on Domestic Violence Ghana:
Health Care Providers
Approach: Observations and in-depth interviews at health facilities (two per region and
>40 staff. Two staff per site). In each urban and rural site, health facilities that provide
emergency care for victims of domestic violence are visited. In each of these sites, both
observations and interviews should take place. Purposive sample of individuals who are
directly involved in domestic violence. These can be leaders or front-line staff or both
depending on the facility. Facilities are equally divided across selected urban and rural
sites.
Facilitator introduces the team, explains consent and the purpose.
Thanks for giving us your time. We very much appreciate your willingness to participate.
My name is [name of facilitator], and I will serve as the interviewer for today’s discussion.
Assisting me is [name of assistant], who will take notes. This study is part of the National
Survey on Domestic Violence 2015 being conducted by the Institute of Development
Studies, University of Sussex (IDS) for the Ministry of Gender, Children and Social
Protection and DfiD, UK. It covers all 10 regions of Ghana, urban and rural areas.
Participation in this research is voluntary. The discussion may last for about 30 to 60
minutes. All information provided will be kept secret. We will not ask your real name
and/or your address. You can withdraw from this interview anytime you want, or refuse
to answer any questions that you are not comfortable with.
The purpose of today’s visit and discussion is to learn from you about:
The causes and consequences of domestic violence, and which services are available
to survivors of domestic violence
What works well, what does not
How these services can be improved or strengthened
The findings will inform policy recommendations to the Ministry of Gender,
Children and Social Protection. It will also enable civil society organisations to reach
out more directly to survivors of domestic violence and lobby for more
programmatic interventions and changes in laws.
There will be a report that will be published.
Facilitator explains the consent forms and the value of recording, and checks
permission.
Facilitator and Note-taker help to complete the consent form.
Note-taker fills out the clip-on sheet.
Facilitator asks for permission to record the discussion.
358
Note to Facilitator: You start with broad opening questions to help us understand some
more detail about the participant. It’s important to learn more about the person’s work;
this will help us to understand what they say, and what they might not feel comfortable
saying. It helps to understand the language that the participant will use.
Discussion starts.
Overview, background: tell me about yourself
1. Can you tell me about the work you do?
Probe:
a. As a health professional, what are some of your main responsibilities?
b. As a health professional, what are some of the main challenges in your work?
c. As a health professional, do you see any problems with the way men treat
women? With the way women treat men? With the way parents treat their
children?
Forms and causes of domestic violence
2. Can you tell me what you know about disagreement and fights by people from the
same family?
3. What kinds of fights do you see? [Probe to find about forms:
a. Are these fights verbal, physical, or sexually related?
b. Who is involved? Probe for: partners, parents and children, adults and non-
biological children?]
4. Why is there fighting/abuse and violence within families?
[Probe to find out about causes: Are they about money, about infidelity, about school
etc.?]
5. Based on your observations – as a professional – how often do you think it happens?
[Probe: At least once a week or once a month?]
6. Are there specific times when this fighting takes place?
[Probe: Any seasons or specific days? Is it linked to school terms? Is it linked to
monthly payment cycles?]
Impact of domestic violence
7. Who is affected by fights in the family? How? Why these people? [Probe: the range
of effects – they include losing confidence, being afraid to leave the house, not being
able to work and becoming trapped in a vicious cycle of dependence, for example.
How are these effects distributed across members of the household, who are
affected by the violence?]
8. Do men also experience violence in the family? What does this look like? How is this
different from how women experience violence in the domestic context?
9. What is the immediate impact? And what is the long-term impact on those affected?
[Probe for economic, psychological, health impacts.]
359
Acceptability of violence and social norms
10. What would make members of a family turn a blind eye to violence?
11. What are the expectations of masculinity (breadwinner, disciplinarian) that can
make men feel they need to behave in a certain dominant or violent way?
12. What are the expectations of femininity (submissiveness) that can make women feel
they need to accept certain dominant or violent ways from men?
13. What would make health officers turn a blind eye to reported cases of domestic
violence? (i.e. discourage referral to police and other institutions for help)
14. Discuss – what are the social pressures – family and community pressure to
encourage or discourage health officers’ referral to the police for prosecution.
15. What are the government pressures to encourage or discourage prosecution of
reported cases of domestic violence?
Intervention and mediation
16. How do these fights come to an end? Who do people go to for help? Who is
involved? [Probe: to find out about government and non-government service
providers:
a. Are these government-appointed people (nurses, social workers, teachers)?
b. Do people feel more comfortable going to community leaders than to police
officers to get support?
c. Do people feel more comfortable going to certain police (men/women,
senior/junior) than other people/service providers? Why?
d. What kinds of things matter in how people make decisions to get support?
What catalyses these decisions?]
17. What kind of services are there to help people? [Probe to get a list of services to
understand the ‘in principle’ facets of government support as the participant
understands this.]
18. Which of these services do people use the most? Why? Which are not useful?
19. What has been your experience with government support for people hurt by other
people in their family? (Discuss satisfaction with police handling of case
reporting/help at health centres.)
20. What do you think needs to happen to:
a. Help those people who commit violence to stop?
b. Help people who experience violence to get support?
c. Help government services reach the right people to end the violence?
d. Which services are seen to be most crucial? Why?
21. How many health cases from domestic violence did you received in this health
facility in:
2012 ……….………………………………………………………………
2013 ……….………………………………………………………………
2014 ……….………………………………………………………………
22. How many of these cases were female victims? How many were male victims
(Probe: ask for rough estimate on average/percentage.)
360
Availability of logistical support
23. Does the facility have specialised departments offer services to victims of violence
such as:
Rape-related services such as post-exposure prophylaxis for HIV
prevention (PEP), sexually transmitted infection (STI) referrals, and
emergency contraceptives (‘morning-after pills’)?
Legal referrals including prosecution advice for victims of violence (Is this
advice also given and tailored to men, and to gay men/women?)
Counselling services, mental health services
Any other (name service)………………………………………………………
Availability of protocols
24. Is domestic violence different from other forms of violence? Does the health unit
have formal protocols to follow when dealing with victims of domestic violence?
Yes
No
If Yes, name the documents …………………………………………………………..
If yes, what do these protocols deal with?
Emergency response to victims of violence
Priority for victims of abuse
Writing of medical reports
Referral to specialised units (paediatrician/social workers/police)
Capacity-building
25. Sensitising the health professionals on domestic violence. Has health centre staff
undergone:
Initial training on gender-based violence and on domestic violence
Continuing training on domestic violence
Specialised in listening skills for victims story
Training in writing medical reports
Affordable health care
26. Are medical reports free?
If No, How much is charged? …………………………………..
Is this a formal, required fee by the government or a personal fee charged by the
doctor? ……………………………………………………………………..
27. How are victims who are unable to pay fee for medical reports able to obtain the
report to forward to police for prosecution of the case?
Continuum of care
28. Does this health site work with any other service providers? If yes, which? (Police,
social workers) Name the organisations here
………………………………………………………………………………………..……..
361
29. How effective is this collaboration?
30. Can you give me an example where this site collaborated with other service
providers to give support to someone who had experienced domestic violence?
[Probe: Who was involved? What did each ‘sector’ do? What were the outcomes for
the person involved? Was the collaboration successful? What made it work well?]
OBSERVATIONS
While waiting look for:
Are files stored away or can the public see these easily? Do you hear staff talk about cases
and private affairs of patients in front of you when you are in a public area? Is there a
private room or other privacy measures for patients when they speak with medical staff?
Ask: Can you show us the building and the services as if we are a victim of domestic
violence who comes to report? Can you show us the process step by step? Where would
we go first, and what would we do? Who would help us? Where would we sit to report?
When and where do I hear about PEP and emergency contraceptives? (Are there rape kits?)
Note: If respondents mentioned special services such as rape-related services, legal
referrals or counselling services, you can ask them to show you where these are.
N.B. These are just guidelines. If any interesting additional issues emerge, the
discussion will also explore these.
Facilitator asks respondent if (s)he has any questions for us.
Facilitator thanks respondent for her/his time and the interesting information they have
shared.
362
3. Observation and In-Depth Interview Guideline on Domestic Violence Ghana:
District Police/DOVVSU Institutions
Approach: Observations and in-depth interviews at police facilities (two per region and
>20 members of the police). In each urban and rural site, police stations are visited; in each
of these sites, both observations and interviews should take place. Purposive sample of
individuals who are directly involved in domestic violence. These can be leaders or front-
line staff equally divided across selected urban and rural sites.
Facilitator introduces the team, explains consent and the purpose.
Thanks for giving us your time. We very much appreciate your willingness to participate.
My name is [name of facilitator], and I will serve as the interviewer for today’s discussion.
Assisting me is [name of assistant], who will take notes. This study is part of the National
Survey on Domestic Violence 2015 being conducted by the Institute of Development
Studies, University of Sussex (IDS) for the Ministry of Gender, Children and Social
Protection and DfiD, UK. It covers all 10 regions of Ghana, urban and rural areas.
Participation in this research is voluntary. The discussion may last for about 30 to 60
minutes. All information provided will be kept secret. We will not ask your real name
and/or your address. You can withdraw from this interview anytime you want, or refuse
to answer any questions that you are not comfortable with.
The purpose of today’s visit and discussion is to learn from you about:
The causes and consequences of domestic violence, and which services are available
to survivors of domestic violence
What works well, what does not
How these services can be improved or strengthened
The findings will inform policy recommendations to the Ministry of Gender,
Children and Social Protection. It will also enable civil society organisations to reach
out more directly to survivors of domestic violence and lobby for more
programmatic interventions and changes in laws.
There will be a report that will be published.
Facilitator explains the consent forms and the value of recording, and checks
permission.
Facilitator and Note-taker help to complete the consent form.
Note-taker fills out the clip-on sheet.
Facilitator asks for permission to record the discussion.
363
Note to Facilitator: You start with broad opening questions to help us understand some
more detail about the participant. It’s important to learn more about the person’s work;
this will help us to understand what they say, and what they might not feel comfortable
saying. It helps to understand the language that the participant will use.
Discussion starts.
Overview, background:
1. Can you tell me about the work you do?
Probes:
a. As a police officer, what are some of your main responsibilities?
b. As a police officer, what are some of the main challenges in your work?
i. As a police officer, do you see any problems with the way men treat
women? With the way women treat men? With the way parents treat
their children?
Forms and causes of domestic violence
2. Can you tell me what you know about disagreement and fights by people from the
same family?
3. What kinds of fights do you see?
[Probe to find about forms:
a. Are these fights verbal, physical, or sexually related?
b. Who is involved? [Probe for: partners, parents and children, adults and non-
biological children.]
4. Why is there fighting/abuse and violence within families?
[Probe to find out about causes. Are they about money, about infidelity, about school
etc.]
5. Based on your observations – as a professional – how often do you think it happens?
[Probe: At least once a week or once a month?]
6. Are there specific times when this fighting takes place?
[Probe: Any seasons or specific days? Is it linked to school terms? Is it linked to
monthly payment cycles?]
Impact of domestic violence
7. Who is affected by fights in the family? How? Why these people? [Probe: the range
of effects – they include losing confidence, being afraid to leave the house, not being
able to work and becoming trapped in a vicious cycle of dependence, for example.
How are these effects distributed across members of the household who are affected
by the violence?]
8. Do men also experience violence in the family? What does this look like? How is
this different from how women experience violence in the domestic context?
9. What is the immediate impact? And what is the long-term impact on those affected?
[Probe for economic, psychological, health impacts, children’s lives.]
364
Acceptability of violence and social norms
10. What would make members of a family turn a blind eye to violence?
11. What are the expectations of masculinity (breadwinner, disciplinarian) that can
make men feel they need to behave in a certain dominant or violent way?
12. What are the expectations of femininity (submissiveness) that can make women feel
they need to accept certain dominant or violent ways from men?
13. Discuss what are social pressures – family and community pressure to encourage or
discourage prosecution.
14. What are the government pressures to encourage or discourage prosecution of
cases?
Intervention and mediation
15. How do these fights come to an end? Who do people go to for help? Who is
involved? [Probe: to find out about government and non-government service
providers:
a. Are these government-appointed people (nurses, social workers, teachers)?
b. Do people feel more comfortable going to community leaders than to police
officers to get support?
c. Do people feel more comfortable going to certain police (men/women,
senior/junior) than other people/service providers? Why?
d. What kinds of things matter in how people make decisions to get support?
What catalyses these decisions?]
16. What kind of services are there to help people? [Probe to get a list of services to
understand the ‘in principle’ facets of government support as the participant
understands this.]
17. Which of these services do people use the most? Why? Which are not useful?
18. What has been your experience with government support for people hurt by other
people in their family? (Discuss satisfaction police handling of case reporting/help
at health centres.)
19. What do you think needs to happen to:
a. Help those people who commit violence to stop?
b. Help people who experience violence to get support?
c. Make sure government services reach the right people to end the violence?
Which services are seen to be most crucial? Why?
Availability of logistical support
20. Does the unit have specialised departments which offer services such as:
Advocacy for victims of violence
Shelter for victims of violence
Continuing Education Units
Counselling for victims of violence
Self-help advice for victims of violence
Protection for victims of violence
Prosecution advice for victims of violence
365
Batterer’s prevention counsellors
Any gender-specific support services (name service ……………………...)
Availability of protocols
21. Does DOVVSU have formal protocols to follow when dealing with reporting of
incidences of domestic violence?
Yes
No
If Yes, name the documents …………………………………………………………
If yes, what do these protocols deal with?
Dispatch response
Arrests/service of orders
Investigation techniques
Evidence
Reports
Follow-ups
Victim’s rights
Rights perpetuators of violence
Emergency judicial response
Refuge/victim’s safety
Capacity-building
22. Have DOVVSU staff received the following specialised training to provide specific
help to the victims of violence?
Initial training on gender-based violence and on domestic violence
Continuing training on domestic violence
Domestic violence and sexual assault training
Specialised in listening skills for victim’s story
Case processing and pace of justice
23. Does DOVVSU/police station have formal procedures to follow to process domestic
violence cases?
If Yes, describe the procedure ..…………………………………………………...
………………………………………………………………………………………..
If No, how are cases of domestic violence processed?
………….…………………………………………………………………………….
How many cases have been reported in
2012
2013
2014
How many cases have been prosecuted?
2012
2013
2014
366
Continuum of care
24. Does this police facility work with any other service providers? If yes, who? (social
workers). Name them
…………....…………………………………………………………………………………..
25. How effective is this collaboration?
26. Can you give me an example of where this facility collaborated with other service
providers to give support to someone who had experienced domestic violence?
OBSERVATIONS
While waiting, look for:
Are files stored away or can the public see these easily? Do you hear staff talk about cases
and private affairs of patients in front of you when you are in a public area? Is there a
private room where victims can file a complaint?
Ask:
Can you show us the building and the services as if we are a victim of domestic violence
who comes to report? Can you show us the process step by step? Where would we go first,
and what would we do? Who would help us? Where would we sit to report?
N.B. These are just guidelines. If any interesting additional issues emerge, the
discussion will also explore these.
Facilitator asks respondent if (s)he has any questions for us.
Facilitator thanks respondent for her/his time and the interesting information they have
shared.
367
4. In-Depth Interview Guideline – Domestic Violence Ghana Opinion/Community
Leaders
Approach: For in-depth interviews with opinion leaders, activists and legal experts: 20
community leaders (e.g. religious leaders, and formal and informal political leaders); 10–
20 legal practitioners; 10–20 activists (national or regional level); and 10–20 journalists
(national or regional level). Purposive sample in all regions of individuals who have been
in current position over two years, equally divided across selected urban and rural sites
alike.
Facilitator introduces the team (2 people maximum) and explains consent and the
purpose.
Thanks for giving us your time. We very much appreciate your willingness to participate.
My name is [name of facilitator], and I will serve as the interviewer for today’s discussion.
Assisting me is [name of assistant], who will take notes. This study is part of the National
Survey on Domestic Violence conducted for the Ministry of Gender, Children and Social
Protection. It covers all 10 regions of Ghana, urban and rural areas. We use both
quantitative and qualitative methodologies.
Participation in this research is voluntary. The discussion may last for about 30 to 60
minutes. All information provided will be kept secret. We will not ask your real name
and/or your address. You can withdraw from this interview anytime you want, or refuse
to answer any questions that you are not comfortable with.
The purpose of today’s discussion is to learn from you about
The causes and consequences of domestic violence, and which services survivors of
domestic violence use
What works well, what does not
How these services can be improved or strengthened
The findings will also inform civil society organisations and help them to reach out
to survivors of domestic violence and lobby for more programmatic interventions
and changes in laws.
There will be a report that will be published internationally.
Facilitator explains consent forms and the value of recording, and checks permission.
Facilitator and Note-taker help to complete the consent form.
Note-taker fills out the clip-on sheet.
Facilitator asks for permission to record the discussion.
368
Note to Facilitator: You start with broad opening questions to help us understand some
more detail about the participant. It’s important to learn more about the person’s work;
this will help us to understand what they say, and what they might not feel comfortable
saying. It helps to understand the language that the participant will use.
Discussion starts.
Overview, background: tell me about yourself
1. Can you tell me about the work you do?
[Probe:
a. As a religious/political/community leader, what are some of your main
responsibilities?
b. As a religious/political/community leader, what are some of the main
challenges in your work?
c. As a religious/political/community leader, do you see any problems with the
way men treat women? With the way women treat men? With the way
parents treat their children?]
Forms and causes of domestic violence
2. Can you tell me what you know about disagreement and fights by people from the
same family?
3. What kinds of fights do you see?
[Probe to find about forms:
a. Are these fights verbal, physical, or sexually related?
b. Who is involved? Probe for: partners, parents and children, adults and non-
biological children]
4. Why is there fighting/abuse and violence within families?
[Probe to find out about causes. Are they about money, about infidelity, about school
etc.?]
5. Based on your observations – as a professional – how often do you think it happens?
[Probe: At least once a week or once a month?]
6. Are there specific times when this fighting takes place?
[Probe: Any seasons or specific days? Is it linked to school terms? Is it linked to
monthly payment cycles?]
Impact of domestic violence
7. Who is affected by fights in the family? How? Why these people? [Probe: the range
of effects: losing confidence, being afraid to leave the house, not being able to work
and becoming trapped in a vicious cycle of dependence, for example. How are these
effects distributed across members of the household who are affected by the
violence?]
8. Do men also experience violence in the family? What does this look like? How is
this different from how women experience violence in the domestic context?
369
9. What is the immediate impact? And what is the long-term impact on those affected?
[Probe for economic, psychological, health impacts.]
Acceptability of violence and social norms
10. What would make members of a family turn a blind eye to violence?
11. What are the expectations of masculinity (breadwinner, disciplinarian) that can
make men feel they need to behave in a certain dominant or violent way?
12. What are the expectations of femininity (submissiveness) that can make women feel
they need to accept certain dominant or violent ways from men?
13. Discuss what are social pressures – family and community pressure to encourage or
discourage prosecution.
14. What are the government pressures to encourage or discourage prosecution of
cases?
Intervention and mediation
15. How do these fights come to an end? Who do people go to for help? Who is
involved?
[Probe: to find out about government and non-government service providers:
a. Are these government-appointed people (nurses, social workers, teachers)?
b. Do people feel more comfortable going to community leaders than to police
officers to get support?
c. Do people feel more comfortable going to certain police (men/women,
senior/junior) than other people/service providers? Why?
d. What kinds of things matter in how people make decisions to get support?
What catalyses these decisions?]
16. What kind of services are there to help victims of domestic violence? (Get a list of
services to understand the ‘in principle’ facets of government support as the
participant understands this.)
17. Which of these services do people use the most? Why? Which are not useful?
18. What has been your experience with government support for people hurt by other
people in their family? (Discuss here satisfaction with police handling of case
reporting/help at health centres.) Are there interventions to prevent domestic
violence in this community?
.……………………………………………………………………………………………….
19. Which of these interventions work against domestic violence? Why?
20. What can the government do better, or differently, to make sure that its policy is
properly implemented?
a. In rural areas?
b. In peri-urban areas?
c. In urban areas?
d. Through: Police? Health workers? Social workers?
21. What do you think needs to happen to:
a. Help those people who commit violence to stop?
b. Help people who experience violence to get support?
370
c. Help government services reach the right people to end the violence?
d. Which services are seen to be most crucial? Why?
N.B. These are just guidelines. If any interesting additional issues emerge, the
discussion will also explore these.
Facilitator asks respondent if (s)he has any questions for us.
Facilitator thanks respondent for her/his time and the interesting information they have
shared.
371
5. In-Depth Interview Guideline – Victims/Survivors of Domestic Violence Ghana
Approach: For in-depth interviews with 45 survivors of domestic violence (three per
enumerator) identified during the quantitative research – the survey. Purposive sample in
all regions of individuals who self-identify as victims of domestic violence, equally divided
across urban and rural areas alike.
Facilitator introduces him or herself, explains consent and the purpose.
Thanks for giving us your time. We very much appreciate your willingness to participate.
My name is [name of facilitator], and I will serve as the interviewer for today’s discussion.
My colleague [name of note-taker], will take notes. This study is being conducted for the
Ministry of Gender, Children and Social Protection. It covers all 10 regions of Ghana, urban
and rural areas.
Your participation in this research is voluntary. The discussion may last for about 30 to 60
minutes. All information provided will be kept secret. We will not ask your real name
and/or your address. You can withdraw from this interview anytime you want, or refuse
to answer any questions that you are not comfortable with.
You were invited for this interview because you participated in a survey during which you
self-identified as someone who has or is experiencing domestic violence. Thank you for
agreeing to speak some more with us about your experiences.
The purpose of today’s discussion is to learn from you about:
the causes and consequences of domestic violence;
which services are available to victims/survivors of domestic violence;
what works well for victims/survivors of domestic violence, and what does not; and
how these services can be improved or strengthened.
There will be a report that will be published from this country-wide study; however,
individual stories will not be identified, so no one will know that you spoke to us or what
you told us.
Facilitator explains about consent forms and the value of recording, and checks
permission. Facilitator helps to complete the consent form.
Discussion starts.
Background information
1. Can you tell me a little bit about yourself? [Interviewer must ask and record the
following background information for each interview: age; sex; marital status
(single/married/living together/living together with children); educational level;
region; district; city/town/rural site of interview.]
372
2. About your home, and who you live with?
3. About your family, who forms part of the house/household (nuclear or extended family
compound household/ rented accommodation etc.); and who do you take care of/takes
care of you in the home?
4. About your daily activities, and what takes up your time during the day (earning a
living or supplementing income/household chores or combination of these etc.)?
5. About growing up, where you grew up and how you came to live here?
Forms and causes of domestic violence
6. You told us that you experienced some violence from other family members of your
household in your life? Is that correct?
7. Can you tell me a little bit more about this experience? [Interviewer should give the
respondent the opportunity to briefly narrate uninterrupted their experience of
domestic violence – notes should be taken of this narration.]
8. Is this situation of violence ongoing? Or was it in the past? When?
9. Let us go into more detail here. What kinds of fights and disagreement did/do you
experience/see in your own family?
[Probe to find about forms: Are these fights verbal, physical, sexual, economic or
emotional/psychological?]
10. Who was/are involved?
[Probe to find out about domestic relations: Partners, parents and children, siblings,
aunts, uncles, cousins, grandparents, stepchildren and non-biological children,
house help.]
11. Why does this fighting/abuse and violence take place in your family?
[Probe to find out about causes: Are they about money, about infidelity, about school
etc.?]
12. How often does this fighting and violence take place?
[Probe for incidence: At least once a week or once a month? Are there specific times
when this fighting takes place? Any seasons or specific days? Is it linked to school
terms? Is it linked to monthly payment cycles?]
Impact of domestic violence
13. Who is affected by fights in your family? How? Why these people?
[Probe for the range of effects – they may include:
a. Victim losing confidence, being afraid to leave the house, not being able to
work and becoming trapped in a vicious cycle of dependence.
b. The effects distributed across members of the household who are affected by
the violence (by age, sex, biological children/foster children/other kin).
c. How are they impacted by the incidences of violence?]
14. What is the immediate impact? And what is the long-term impact on those affected?
[Probe for short-/long-term impacts: economic, psychological, health impacts.]
Victim responses to violence
15. When you and your family have conflicts, what do you do to cope with the violence?
373
16. When a person is angry or uses violence or threatens to use violence against you, have
you ever sought support from other people in the family?
Intervention and mediation
17. Does anyone in the family come to intervene/help you in any situations of family
violence? Who? How? Why?
18. Is there any one to help reconcile the conflicts between you and the person who has
hurt you in the family? Please explain.
[Probe for family/extended family intervention and mediation.]
19. Have you been helped by neighbours or local authorities/local organisations (i.e. chief
or queen mother, women’s union, reconciliation group)? Please explain the reasons
why you have or have not sought help from these people/organisations.
[Probe: Community intervention and mediation.
a. If no, why did you not seek help from these people community facilities?
b. If yes, who/which did you go to for assistance? What were their responses
when you told them your story or asked them for help? How did these
help/support you? Were you satisfied with the help they provided? Did their
support help end the violence you were facing from the family member?]
Knowledge/availability of victim support services
20. Can you tell us what you know to do/have been advised to do when you are hurt by
members of your family?
[Probe for knowledge of rights/resources:
a. Do you know about the Ghana 2007 Domestic Violence Act 732/gender
equity laws?
b. Do you know about any domestic violence victim support services (ask
specifically about DOVVSU, CHRAJ, others)?]
21. When you were injured from the violence, did you go to the police for help?
[Probe for use/availability of support services and programmes:
a. If no, why did you not use the police facilities?
b. If yes, how did the police facility help/support you?
c. Were you satisfied with the help they provided?
d. Did the police support help end the violence you were facing from the family
member?]
22. When you were injured from the violence, did you go to health facilities such as
hospital, commune health station, pharmacy stores?
[Probe:
a. If no, why did you not use these facilities?
b. If yes, which of the health facilities did you go to (public or private)?
c. How did the health providers help/support you?
d. Did you have to pay for the services?
e. If yes, did you have enough money to pay for the services?
f. Did you have someone to accompany you, or did you go by yourself? If yes,
who?]
374
23. What do you think local leaders can do to prevent/support victims of domestic violence
(to find out about the role of chiefs, community leaders, local assembly/MP and other
administrative/political officials play in implementing policy directives)?
24. What do you think the police facilities can do to support victims of domestic violence
(to find out about police support and help to prevent domestic violence)?
25. What do you think the health facilities can do to support victims of domestic violence
(to find out about health support for victims and help to prevent domestic violence)?
26. What do you think are the most effective ways to stop other people from being hurt
like this?
[Probe:
a. Are these types of support available? Where do you think you can find them?
b. How does/did the absence of any of these types of support influence your
situation?]
27. Looking back at your situation, what advice would you give another woman or man,
or child, who has just started to have these sorts of problems in their home?
[Probe: What should these people do? Ask about all the options/choices they have or
do not have]
28. How do you think people who hurt other people, like you have been hurt, should be
punished? What should happen?
[Probe: How do people think about justice, and appropriate punishment?]
29. Are you a member of a group or an organisation? Which (religious, social or traditional
group, saving and credit or any group)?
30. If you could advise the government, what would you tell it to do? [This is an open-
ended question that can be really useful in drawing out unique responses about
government accountability to end domestic violence, so encourage respondent.]
N.B. These are guidelines. If any interesting additional issues emerge, the discussion
should also explore these.
WRAP UP
Facilitator asks respondent if (s)he has any questions for interviewing team.
Facilitator thanks respondent for her/his time. (Thank you for sharing this with us. We
have asked very difficult questions, and thank you for being so open. All shared
information will remain confidential. What you have told us is very important, and will
help us in our work to address domestic violence.)
From the victim’s responses, mention his/her strengths.
Give details of follow-up counselling support available both immediately and later.
Give more general information about services available in the community.